mirror of
https://github.com/rennf93/roboco.git
synced 2026-08-03 07:23:24 +02:00
7c8453e21049f884ec9bbe8d782668c14fc8dae2
530
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 Renn FandRenzo F
|
3838d64eaa |
sandbox: kitchen-sink images, feature-aware selection (Phase 2)
Phase 1 made the provisioner able to activate allowlisted extensions post-ready but kept the bare upstream images. Phase 2 ships the images that actually carry the extension/module files, and selects them only when a venture requests features — bare sandboxes stay on the light upstream image (no heavier pull, honoring the 'existing opters stay bare' decision). - _PostgresEngine / _RedisEngine gain kitchen_sink_image + image_for(features): bare (no features) -> the light image; features requested -> the kitchen-sink image. The provisioner runs engine.image_for(features), not engine.image, so the bare path is byte-for-byte unchanged. Mongo inherits the base image_for (returns its image regardless — no activatable features). - docker/sandbox-pg.Dockerfile: pgvector/pgvector:pg16 (ships vector) + postgis apt install; contrib (pg_trgm/citext/uuid-ossp) inherited from the official postgres base. Built at deploy via the sandbox-pg-image compose one-shot (mirrors the agent-image builders); the provisioner's _ensure_image finds the local tag and never pulls. Published by release.yml; pulled in registry compose. The verify step fails loudly if an extension's files are missing. - _RedisEngine kitchen-sink image: redis/redis-stack-server:latest (headless; ships search/json/bloom as loadable-but-unloaded modules — no custom build). - Extended the sandbox image-tag ghost-tag guard (the mongo:8-alpine regression test) to also cover kitchen_sink_image: skips locally-built roboco-* images, uses the namespaced Docker Hub endpoint for redis/redis-stack-server. Image-specific package names / module .so paths are verified at the CEO's NAS deploy (the spec's NAS smoke); the unit tests with the fake runner remain the CI bar, and the verify step is the fail-loud safety net for a wrong build. |
||
|
Renn FandRenzo F
|
b015cde9ad |
sandbox: post-ready extension/module activation + allowlist (Phase 1)
Parameterized sandbox dev DBs — groundwork for 'extensions on the fly'
(docs/internal/specs/2026-07-13-sandbox-extensions-on-the-fly.md). A
venture declares the extensions/modules it needs; the provisioner activates
them post-ready via docker exec, never via bind-mounts or initdb scripts.
Phase 1 (behavior-preserving scaffolding — no image, no schema, no caller
passes features yet):
- Allowlists SANDBOX_PG_EXTENSIONS / SANDBOX_REDIS_MODULES are the ONLY
extensions/modules the system will ever activate — the security
containment, not privilege. plpython3u & co. (superuser-RCE vectors)
are excluded by construction.
- SandboxEngine ABC gains enable_step / verify_step / verify_ok. pg:
CREATE EXTENSION IF NOT EXISTS via psql, verified by a pg_extension
count. redis: MODULE LOAD per module, verified by MODULE LIST. mongo:
no-op (server is batteries-included).
- SandboxProvisioner.provision takes features={service: [names]},
allowlist-validates before any container runs, runs enable then verify
after the base readiness probe; a failed enable or a short verify (image
missing the extension files) is fatal — an agent never receives creds
for a db missing what it asked for. Empty features = bare = the
existing path, byte-for-byte unchanged.
- SandboxConnection gains features; as_payload surfaces
available_extensions / available_modules so the agent doesn't guess.
13 new unit tests (fake docker runner): enable/verify argv per engine,
allowlist rejection of plpython3u before any run, failed-enable + failed-
verify fatality, bare-provision unchanged, payload surfacing.
|
||
|
|
a3524da5f8 |
[90c9474c] Auditor revival: scheduled audit trigger and reactive alert producers (#499)
* [927e64d5] Backend slice: auditor scheduled trigger and reactive alert producers (#496) * [1f2cdb4b] Reactive alert producers at QA-fail and rework (#492) * [1f2cdb4b] feat(services): add auditor-targeted rework alert producers at QA-fail and rework chokepoints * [1f2cdb4b] test(services): fix mypy typing in auditor alert producer unit tests * [1f2cdb4b] docs(backend): document reactive auditor rework alert producers in map and role docs --------- Co-authored-by: Backend Developer 2 <be-dev-2@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [5173415f] Scheduled audit trigger, config, and sweep prompt (#493) * [5173415f] Add scheduled audit trigger, interval config, sweep prompt, and focused tests * [5173415f] Allow ROBOCO_AUDIT_INTERVAL_SECONDS=0 to disable scheduled sweeps * [5173415f] docs(audit): document scheduled auditor sweeps and ROBOCO_AUDIT_INTERVAL_SECONDS --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [a26c18b9] E2E smoke test for auditor triggers (#495) * [a26c18b9] Add e2e smoke test for auditor scheduled and reactive triggers * [a26c18b9] docs(tests): add e2e smoke test catalog and changelog entry for auditor triggers --------- Co-authored-by: Backend Developer 2 <be-dev-2@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [3bc47cdc] Fix _fresh_orchestrator state for auditor trigger e2e tests (#497) * [3bc47cdc] fix(tests): initialize orchestrator state in _fresh_orchestrator helper * [3bc47cdc] docs(changelog): add _fresh_orchestrator test harness fix entry --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [8323cd50] Fix e2e smoke regression on assembled cell PR #496 (#498) * [8323cd50] fix(e2e_smoke): repair auditor-trigger smoke tests and harden harness * [8323cd50] docs(tests): document e2e smoke harness hardening for PR #498 --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> --------- Co-authored-by: Backend Developer 2 <be-dev-2@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> * [37e6d999] Backend: repair failing CI checks on auditor revival PR #499 (#503) * [6e79bada] Triage and fix Python quality gate and Analyze (python) failures (#501) * [6e79bada] fix(task): replace type ignore with forward-reference cast for SQLAlchemy Mapped UUID in get_all_descendants * [6e79bada] fix(notification_delivery): add generic type arguments to dict return types in get_ack_status and get_delivery_summary * [6e79bada] docs(changelog): add Python quality gate type-hygiene fixes to Unreleased --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [50e7e104] Triage Analyze (javascript-typescript) failure on backend-only diff (#500) * [50e7e104] Split CodeQL workflow so JS/TS analyzer only runs on panel changes * [50e7e104] docs(backend): document split CodeQL workflow triggers and branch protection notes --------- Co-authored-by: Backend Developer 2 <be-dev-2@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [203c426b] Triage and fix e2e lifecycle smoke (scripted agents) failure (#502) * [203c426b] fix(orchestrator): pre-initialize _instances in __new__ so __init__-bypass tests survive _dispatch_audit_work; allow audit_interval_seconds=0; mount /api/notifications in e2e harness * [203c426b] fix(e2e_smoke): restore ROBOCO_AGENT_TOKEN isolation and clarify /api/notifications mount comment * [203c426b] docs(map): document orchestrator __new__ pre-init and e2e harness token isolation for auditor-revival smoke fix --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [48cb05c2] Fix remaining e2e lifecycle smoke (scripted agents) failure on auditor-revival PR #503 (#504) * [48cb05c2] Harden AgentOrchestrator __new__ pre-init for auditor dispatch state * [48cb05c2] Document auditor-dispatch pre-init rationale in AgentOrchestrator __new__ * [48cb05c2] docs(orchestrator): extend __new__ pre-init docs for auditor-dispatch state --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> Co-authored-by: Backend Developer 2 <be-dev-2@roboco.tech> * [90c9474c] intake: ambient workspace note + dedupe scope clones by git_url Two intake follow-ups folded into 90c9474c's spec Notes: (a) _resolve_intake_ambient now prepends a workspace note so the intake agent knows its cwd holds clones of every project in the scope (the primary at cwd, siblings alongside under /data/workspaces) and drafts against the real trees via Grep/Glob/Read, not from memory. (b) _clone_intake_scope dedupes slugs by git_url before cloning. A multi-project scope can list several projects pointing at one repo (a monorepo's cell-projects share a git_url); cloning each produced redundant identical workspaces. Mirrors CI-watch's per-git_url dedupe: keep the first slug per non-empty git_url; a project with no/empty git_url is never collapsed onto another so distinct local repos still clone. The dedupe is a pure static helper (_dedupe_slugs_by_git_url) with unit coverage. --------- Co-authored-by: Backend Developer 2 <be-dev-2@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
Renn FandRenzo F
|
69271f9e98 |
fix(task): push a pre-set branch_name when the ref is missing on origin
A branch_name set on a task was treated as proof the ref existed on origin, so _finalize_claim skipped _ensure_branch_for_task and create_branch/push never ran. A manual field write (or a prior failed create_branch whose rollback didn't restore branch_name) left the field set while the branch was never pushed; descendants then ls-remote'd the name, found it empty, and cut from master via create_branch's silent fallback — breaking the cell->root branch hierarchy (MegaTask f7d0a61a root-branch 404). Defect A: - _ensure_branch_for_task trust-but-verifies a pre-set branch_name: probe origin, and when the ref is confirmed missing run the full create to push it. An inconclusive probe (network error) fails soft so a transient glitch can't fail a normal resume claim. Gated on project_id so branchless coordination/umbrella tasks are untouched. - _finalize_claim always runs _ensure_branch_for_task (the single chokepoint that ensures the branch exists) and snapshots+restores branch_name on rollback, so a failed first attempt can't leave the field half-set and short-circuit a retry. - GitService.branch_exists_on_remote: ls-remote probe returning True (present) / False (absent) / None (probe errored, fail soft). |
||
|
|
ba7135ba50 |
feat(gateway): carry intake technical depth down the chain + widen review coherence scope (#491)
Two structural issues flagged by the CEO:
1. Task technical-depth dilution — intake's rich analysis (file:line
targets, code examples, rationale) was getting lost as it traveled
umbrella -> root-subtask -> cell -> dev. The detail IS preserved in
Task.description; the dilution was in delegation (PMs re-authoring)
and the intake prompt not demanding depth.
Fixes:
- evidence_repo: ancestor_context_for_task walks the parent chain
(cycle-guarded, depth-capped 16, desc-clipped 1500) and surfaces it
as parent_context in the evidence payload, so a leaf dev finally
sees the upstream intake analysis instead of a bare title.
- evidence_builder: Task.description now rides in the payload;
EvidencePayload gains description + parent_context (omit-when-empty
so no null noise).
- orchestrator: _description_body (capped 4000) injects the
description into the dev spawn prompt + SessionStart briefing.
- role prompts (main_pm/cell_pm/developer/prompter): teach pass-the-
torch, don't-dim-it; prompter now demands file:line/code-examples
in the_work/notes (reconciled with the no-code-level-ACs-on-roots
rule). main_pm's brief-not-a-spec scoped: not-a-spec applies to the
solution only, facts forward verbatim.
2. PR-review/QA scope too narrow — they only checked the AC checklist,
not whether the change is coherent with project structure/intent.
Fixes:
- qa.md + pr_reviewer.md: Coherence & intent rule (intent via
description+parent_context, coherence with project patterns,
standards). Criterion-less major findings allowed for intent drift
(Finding.criterion is optional).
- parent_context + description wired into the gate/QA/inbound-PR
evidence builders (fail-open, logged).
Skipped per YAGNI: a technical_spec JSONB column (detail is already in
description) and a criterion_kind enum (criterion is already optional).
All gates green: ruff, mypy (1152), pytest (12883 passed, 94.82% cov),
xenon, vulture, bandit, pip-audit, deptry, alembic, import-linter,
foundation-check.
Co-authored-by: Renn F <rennf93@users.noreply.github.com>
|
||
|
|
1114ee5ea0 |
[77719d3f] A2A team telemetry: coordination event notifications for 5 event types (#477)
* [13d03d5c] Add 5 coordination-event notification producers + wire at chokepoints (#472) (#474) * [13d03d5c] Add 5 coordination-event notification producer methods * [13d03d5c] Wire reassignment/collision/unblock/dependency-revival notifications * [13d03d5c] Wire stale-claim-reaped notification into orchestrator reaper * [13d03d5c] fix(runtime): guard reaper's UUID annotation + defensive attr access The stale-claim-reaped notification hook added a runtime-unquoted `UUID` type annotation (only imported under TYPE_CHECKING, so the module raised NameError on import) and a direct `t.assigned_to` attribute access that crashes against the minimal test doubles the existing reaper test suite uses. Quote the annotation and switch to getattr-defensive access, matching `_assignee_is_provider_parked`'s existing convention in the same file. * [13d03d5c] test(notification): unit coverage for 5 coordination-event producers One test per new send_* method (reassignment, collision-sequencing, unblock, dependency-revival, stale-claim-reaped) following the existing _FakeDb/_patch_db_context pattern, asserting subject/body/ related_task_id/priority/recipient-count, plus a no-recipients no-op case for reassignment. * [13d03d5c] test(task): prove reassign + unblock don't double-fire notifications Two chokepoint-level tests mocking NotificationService at its defining module: a repeated reassign() to the same already-current target skips the notification (guarded by comparing against the pre-mutation assignee), and a repeated unblock() on the same task only notifies once since the second call short-circuits on the status!=BLOCKED guard. * [13d03d5c] style(task): ruff format the collision-sequencing wiring block No behavior change — reflows the newly-added _notify_collision_sequencing call site to satisfy ruff format's line-length rules. * [13d03d5c] docs(backend): add coordination-event notification producers guide Documented the 5 new NotificationService producers (reassignment, collision-sequencing, unblock, dependency-revival, stale-claim-reaped) with fire conditions, double-fire prevention mechanisms, and implementation patterns. Updated backend README to link the new services guide for developers integrating new coordination events. --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [3ee8150b] Frontend: render coordination-event notifications + e2e smoke coverage (#475) * [69777c3a] test(e2e-smoke): add coverage for soft-block + unblock coordination notifications (#471) Co-authored-by: Frontend Developer 2 <fe-dev-2@roboco.tech> * [8eb82639] Render 5 coordination-event notification types with task deep-links (#470) * [8eb82639] feat(notifications): add APPROVAL type icon and deep-link component test Add missing APPROVAL member to the frontend NotificationType enum to match backend roboco/models/base.py, wire its icon into the existing typeIcons Record in the notifications page, and add a component test covering type rendering and the task deep-link. * [8eb82639] docs(notifications): document 5 coordination-event types and APPROVAL enum addition Added comprehensive reference guide explaining the 5 notification types (TASK_ASSIGNMENT, BLOCKER_ESCALATION, REVIEW_REQUEST, DOCUMENTATION_REQUEST, APPROVAL), their visual identities (icon + color), use cases, and deep-linking behavior to related tasks. Updated panel README with quick reference table. TypeScript Record pattern ensures exhaustive type coverage at build time. --------- Co-authored-by: Frontend Developer 1 <fe-dev-1@roboco.tech> Co-authored-by: Frontend Documenter <fe-doc@roboco.tech> --------- Co-authored-by: Frontend Developer 2 <fe-dev-2@roboco.tech> Co-authored-by: Frontend Developer 1 <fe-dev-1@roboco.tech> Co-authored-by: Frontend Documenter <fe-doc@roboco.tech> * [a27de2a8] fix(docs): reflow hard-wrapped notification-types.md to pass markdown gate (#479) (#481) The Python quality gate on assembled PR #477 was red because the newly added docs/frontend/components/notification-types.md (introduced by the frontend coordination-event rendering commit) had manually wrapped prose paragraphs, which scripts/reflow_md.py --check rejects as part of make quality. Reflowed the file with scripts/reflow_md.py --apply (whitespace only, no content change) so the check passes. ruff format/check, mypy, xenon, vulture, bandit, and the full pytest suite (10284 passed) all confirmed green on this commit; notification.py, task.py, and orchestrator.py are untouched. Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> * [705419d5] Remove duplicate unblock notification and fix its dependent tests (#485) (#488) * [705419d5] fix(notifications): remove duplicate unblock notification, fix its tests The /unblock route was still calling delivery.notify_assignee_of_unblock() (TASK_ASSIGNMENT) after TaskService.unblock() already sent the send_unblock_notification() ALERT wired in by an earlier task — a real duplicate notification on every unblock. Delete the route-layer call and the now-dead NotificationDeliveryService.notify_assignee_of_unblock method, fix the integration test that mocked it, and fix/extend the e2e notification-coordination-events test to assert the persisted ALERT rows (exact subjects) for both the direct-unblock and dependency-revival producers instead of the old TASK_ASSIGNMENT assertion. * [705419d5] docs(backend): update coordination-events doc for unblock duplicate removal --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [6c142a73] docs(changelog): document restored coordination-event notification producers and add collision-sequencing double-fire test (#489) (#490) Co-authored-by: Backend Developer 2 <be-dev-2@roboco.tech> * [77719d3f] Seed system agent in e2e harness to fix unblock/dependency-revival notifications The e2e harness's seed_company omitted the system sentinel agent that production seeds via initial_data.py. The unblock and dependency-revival notification producers default to from_agent="system", which _resolve_agent_uuid looks up by slug in the DB. With no system row the resolver returns None and _create_notification silently skips the notification, so the two ALERT assertions got 0 rows instead of 1. The soft-block test passed because it uses NotificationDeliveryService which creates the notification directly with a real agent UUID as from_agent, bypassing the slug resolution path entirely. * [77719d3f] Use foundation UUID for system agent to avoid slug collision The first attempt seeded the system agent with a random UUID. Other tests (_seed_system_and_secretary, _seed_video_agents) check by the fixed foundation UUID via session.get(AgentTable, uuid); not finding it they INSERT their own system row, hitting ix_agents_slug. Using the foundation UUID makes their check find the seed_company row and skip. * [77719d3f] Fix dependency-revival notification event loop mismatch The dependency-revival test calls _unblock_dependents directly via stack.run_db, which creates a new asyncio event loop. Inside, _notify_dependency_revival -> NotificationService._create_notification opened its own session via get_db_context(), which reuses the singleton _DbHolder engine — bound to the FastAPI server's event loop. The asyncpg connection raised 'Future attached to a different loop' and the exception was silently caught + logged as a warning, so the notification never persisted and the test saw 0 rows. Fix: add an optional db_session parameter to _create_notification and the two send methods. When provided, use the caller's session directly and skip the internal commit (the caller owns the transaction). The TaskService's _notify_unblock and _notify_dependency_revival now pass self.session, keeping the notification in the same event loop + session as the task transition. * [77719d3f] Scope system-agent seeding to notification tests only Seeding the system sentinel in seed_company (commits 3bba7b32/617b7890) fixed the 0-notification bug but caused 3 i_documented gateway_timeout failures: every e2e test now paid notification-creation latency for system-origin notifications that were previously silently skipped, pushing the already-slow i_documented verb past its 120s timeout. Move system-agent seeding out of seed_company and into a scoped _seed_system_agent helper called only by the two coordination-event tests that exercise send_unblock_notification / send_dependency_revival_notification (both resolve from_agent='system' via DB lookup). dev_lifecycle and state_machine tests revert to the pre-fix behavior (system-origin notifications silently skipped, no extra latency). The event-loop fix (commit |
||
|
|
cea3e56628 |
feat(lifecycle): revision findings ledger — structured failure feedback, persisted and delivered down the chain (#486)
* feat(lifecycle): revision findings ledger — structured QA/PR/PM/CEO failure feedback, persisted and delivered down the chain Every bounce used to survive only as flattened prose: rounds overwrote each other in notes_structured, request_changes persisted nothing, two raw dev_notes appends were silently destroyed by the next handoff note, and the dev prompt pointed at fields (qa_notes via evidence(), pm_notes) the API never delivered. Agents re-interpreted and re-discovered every failure before they could start fixing it. - task_review_findings (migration 071, append-only): file/line/severity/ criterion(AC-id-validated)/expected/actual/fix/evidence per finding, with origin (qa|pr_gate|pm|ceo), round, and an open->addressed->verified lifecycle (waived reserved); new tasks.pm_notes + PmReviewContent give request_changes a structured home - producers: fail_review/pr_fail/request_changes take findings=[...] (prose issues shimmed+merged for one release, deprecation-logged); ceo_reject validates its reason (no 500), lands an origin=ceo finding, and bumps round+audit on branchless coordination roots; guardrails at the verb chokepoint (nudge >5, hard reject >10, field caps, traversal-safe file); the dev_notes data-loss appends are removed; new task.request_changes + task.ceo_reject audit events close rework attribution - delivery: qa_notes/pr_reviewer_notes/pm_notes carry the deterministic [F-id8] rendering; claim briefings, evidence(), the REVISION_REQUIRED spawn prompt, PM triage bounced-blocks, and A2A bodies deliver open findings; round-N+1 QA and gate reviewers get the full prior ledger; panel Findings tab + bounced-xN chip; metrics pm_rejects/ceo_rejects + findings counts; vault task notes render a Findings section (fail-open) - resolution closes for every origin: i_am_done and submit_up/submit_root take resolved_findings gated by FINDINGS_ADDRESSED (owner-gated so a stale non-owner PM can never mutate the ledger); pass_review/pr_pass/ complete verify-stamp same-transaction; ceo_approve stamps best-effort - 24 real-DB integration tests drive the full loop through the real choreographer; full suite 12856 green * docs: revision findings ledger sweep — CLAUDE.md, map, RAG corpus - CLAUDE.md: new ledger section + corrected request_changes row - docs/map/review-findings.md (new subsystem map) + surgical updates to task-service/pr-gate-review/metrics-observability/vault/panel maps - docs/rag: producers' findings contract across qa/pr-reviewer/developer/ cell-pm/main-pm/ceo role docs (the PM docs were missing request_changes entirely), verb references, and a new architecture/review-findings.md disambiguating ledger findings from convention findings * test(e2e): resubmit resolves the pr_fail finding per the ledger contract The scripted pr_fail revision loop resubmitted submit_up without resolved_findings — correctly rejected now that FINDINGS_ADDRESSED gates the PM resubmit verbs (green locally, red only in CI since the e2e suite skips without ROBOCO_E2E_SMOKE=1). The scripted PM now reads the open ledger row pr_fail persisted (new open_finding_ids arc helper) and resolves it on resubmit, asserting the open set drains — exercising the coordinator half of the new contract end to end. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
d03181ab48 |
feat(vault): Obsidian vault V2 — janitor, archival, weekly report, KB ingest, Bases + sync runbook (#482)
* feat(vault): V2 — create-seam + drift janitor, archival, weekly org-report, KB ingest, Bases views + sync runbook Implements the vault V2 canonical spec end to end (the splice guard shipped separately and is reused at KB-ingest time): - materialize-on-create: TaskService.create writes each task's note best-effort from the moment it exists; the transition-touch stops no-oping on live work - drift janitor (services/vault_janitor.py + hourly _vault_janitor_loop): daily changed-task re-projection, random drift sample, archival pass — restart-proof via RoboCo/_meta/.janitor_state.json, 200/cycle caps, per-item isolation, processed-only resume markers, self-repairing state file - archival: vault_archive_days (30, 0=off) moves old terminal tasks' notes to RoboCo/Archive/<year>/Tasks/<project>/ — one write_task code path for janitor and rebuild, id8 lookup across Tasks/+Archive/, alias links keep moves safe - weekly org-report: VaultWriter.write_org_report renders Reports/<ISO-week>.md from MetricsService/UsageService (numbers duplicated into frontmatter for trend queries), once per ISO week, with a best-effort CEO notification - KB ingest: IndexType.VAULT_NOTES + VaultNotesIndexPlugin + _vault_kb_loop embed the CEO's RoboCo/Notes into the RAG corpus — injection guard as a hard gate (flagged notes quarantined with an idempotent callout), traversal- and symlink-contained at both config and engine layers, content-hash dedup, 50-ingest/cycle cap, frontmatter stripped; reaches roboco_kb_search, the mentor default domain, claim-time briefings (kind vault_note), and the panel KB browser; no migration (chunks table auto-creates; migration 030's CHUNK_TABLES tuple appended per the chunks_playbooks precedent) - Bases views (Task Board.base, Reports.base — schema verified against the Obsidian docs) + the Mac sync runbook vault asset - config/flags/compose: vault_archive_days, vault_report_enabled (flags card), vault_kb_enabled (flags card; NAS compose arms it, registry ships it off), vault_kb_dirs (+ overlap/traversal validator), vault_kb_interval_seconds - e2e smoke (tests/e2e_smoke/test_vault_v2.py): real create-seam, real janitor cycle incl. archival + state, real KB engine + real guard * docs: vault V2 sweep — map, RAG corpus, CLAUDE.md - docs/map/vault.md: V1+V2 — janitor/archival/report/KB data flows, new files, config, health posture - docs/map/orchestrator.md + task-service.md: the two new loops, the create seam, the three janitor queries - docs/rag/architecture/obsidian-vault.md: agent-facing what-changed (notes from creation, archive link-safety, CEO notes retrievable, weekly report) - docs/rag/architecture/config-reference.md: the five new settings - CLAUDE.md: vault paragraph covers V1+V2; flags-card list mentions the vault report/KB flags --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
f2834cf521 |
fix(tasks): merge cross-lineage dependency content at branch cut (#466)
* fix(tasks): merge cross-lineage dependency content at branch cut The dependency gate enforced timing but never content: a dependent's fresh branch could miss a same-repo dependency's merged work when that merge landed outside the branch's own ancestor chain (cross-cell edges under one root, same-repo batch cross-root edges). After a successful branch cut, each dependency's real merge target (resolve_parent_branch) is fetched and, unless already an ancestor, merged into the new branch; conflicts abort cleanly (branch stays at its cut point, warning + an accumulating task marker note) and never fail the claim. Cross-repo dependencies are skipped — no shared history. Zero git work for the no-deps common case; resumes never re-enter (branch creation only). * chore(foundation): regenerate lifecycle artifacts; reflow inherited prose * [lineage] mypy-clean mock idioms in the lineage orchestration tests --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
50ec283533 |
fix(api): settings PUT accepts booleans/numbers from the panel (#465)
* fix(api): settings PUT accepts the JSON scalars the panel sends The feature-flags card sends booleans and numeric settings send numbers; SettingUpdate.value was typed str, so pydantic 422'd on type before the per-key validators ever ran (live: PUT /settings/notifications_enabled). Scalars now coerce to the stored text form — bools to the 'true'/'false' the validators parse. * chore(docs): reflow hard-wrapped prose from the #401 merge * chore(foundation): regenerate lifecycle artifacts; reflow inherited prose --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
53a028ec04 |
fix(agents): raise tool-call budget to 300 (halt) / 100 (warn) (#464)
* fix(agents): raise the tool-call budget — 150 halted legitimate work mid-task Repeated budget-sweep bounces: a dev at the 150 ceiling gets its container killed seconds after a real commit, burning a spawn and the resumed agent's re-verification turns. 300 keeps the runaway guard while clearing a real task's footprint; warn scales to 100. * chore(docs): reflow hard-wrapped prose from the #401 merge * chore(foundation): regenerate lifecycle artifacts; reflow inherited prose --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
20110debab |
fix(ci): fleet-branch push triggers + dispatcher claim prefilter (#463)
* fix(ci): fleet-branch push triggers close the absent-check gap; dispatcher claim prefilter PROVEN with API receipts: when the PM squash-merges a subtask PR into a branch that is itself another PR's head (GitService.merge_pull_request → GitHub's Merge API), the pull_request synchronize webhook fires unreliably (1 of 3 in the live sample) while plain push events fired 100% — so PR heads sat with ABSENT required checks that three review rounds mistook for green. CI, CodeQL, e2e-smoke, and panel-ci now also trigger on push to the fleet's branch types, deduped by a concurrency group keyed on head_ref||ref_name so a branch that is also a PR head never double-runs. Dispatcher churn: _route_unassigned_pm_task consults the claim guards' own predicate (TaskService.is_pending_claim_blocked, a public wrapper — no duplicated SQL) before routing, so dependency- or sequence-held tasks skip the tick with zero HTTP claim round-trips; fails open so a DB hiccup degrades to the old behavior. * chore(docs): reflow hard-wrapped prose inherited from the six-PR merge train * chore(foundation): regenerate lifecycle artifacts; reflow inherited prose --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
786e6ffc3c |
fix(security): prompt-injection screening for engine-ingested external text (#462)
* fix(security): screen engine-ingested external text for prompt injection The X mentions poll and the vault inbox both fed attacker-writable text (tweets, tagged notes incl. meeting-bridge output) raw into local-model prompts and CEO-facing draft payloads. The agent-sdk prompt guard's detection moves to a pure roboco/foundation/policy/injection_guard.py (prompt_guard re-exports it — grok path byte-identical) and gains screen_external_text: per-line detection where a matched line is flagged in place, never dropped, and the whole text rides an explicit untrusted-content envelope. Both engines screen once at ingestion and use the screened rendering for the model prompt AND the persisted marker/description — including the vault engine's deterministic LLM-failure fallback, which previously used the raw body verbatim. * chore(docs): reflow hard-wrapped prose inherited from the six-PR merge train * chore(foundation): regenerate lifecycle artifacts; reflow inherited prose --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
f0f09b2204 |
[1197c975] Re-add Playwright chromium to FE/UX QA images and add browser-verification prompt guidance (#406)
* [3e552255] Re-add Playwright chromium to QA images + prompt guidance (#395) (#405)
* [3e552255] feat(docker): re-add Playwright chromium-headless-shell to QA images
* [3e552255] ci(docker): add Playwright QA image build + headless smoke check workflow
* [3e552255] fix(ci): scope agent-image-smoke.yml trigger to paths only, add PR comment
The workflow was gated by `branches: [master]` on both push and
pull_request, but this repo's task-hierarchy PRs open against nested
parent feature branches, not master, until root->master assembly - so
the workflow never fired on a dev-level PR and produced zero evidence.
Drop the branch filter (path scoping is sufficient) and post the
size-delta table + smoke-check output as a PR comment via
actions/github-script, since no agent role has gh CLI or GitHub API
read access to pull check-run output directly.
* [3e552255] fix(ci): post agent-image-smoke PR comment even on step failure
The 'Post results as a PR comment' step only had
`if: github.event_name == 'pull_request'`, which GitHub implicitly ANDs
with success() — so if the docker build or headless-launch smoke check
failed, the PR comment (the only evidence-delivery path QA/PM has, since
no agent role can read the Checks tab) silently never posted. Added
always() so a partial report always lands on the PR.
This commit also re-lands the branches-filter removal + PR-comment step
from
|
||
|
|
7f138d3bf5 |
[e4ed92d6] Video pipeline per-project requests, re-render action, composition preview (#403)
* [7f2c881a] Project-scope video pipeline + re-render + preview proxy (#386) (#396) * [7f2c881a] feat(video): scope on-demand video requests + render loop to project_id Require project_id on VideoRequestBody (404 when unresolvable or not opted into the video engine), thread it through VideoEngine.open_video_task via a shared resolve_authoring_project helper, and resolve the render loop's motion/ workspace from the authoring task's own project_id instead of the hardcoded self_heal_project_slug. * [7f2c881a] fix(video): cast task.id to UUID before VideoEngine.rerender calls mypy flagged task.id as sqlalchemy.sql.sqltypes.UUID[Any] rather than uuid.UUID in the three rerender tests; cast to UUID per the codebase's established idiom (cast("UUID", obj.id)) used elsewhere for the same SQLAlchemy Mapped-attribute inference gap. * [7f2c881a] docs(video): API endpoints for project-scoped requests, re-render, and preview proxy Add comprehensive API documentation for the new project-scoped video engine endpoints: - POST /api/video/request: on-demand video authoring scoped to project_id (breaking change) - POST /api/video/pipeline/{task_id}/rerender: CEO-triggered re-render with idempotency key clearing - GET /api/video/preview/{task_id}/{file_path}: CEO preview proxy with path-traversal confinement Document project-scoping architecture: authoring tasks and render loop now resolve from task's own project_id instead of hardcoded self_heal_project_slug. Add migration guide covering breaking change to VideoRequestBody schema (project_id now required), error handling changes (404 on unresolvable/non-opted-in projects), and client migration steps. --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [8f959c3b] docs(ux_ui): add project picker, re-render control, and composition preview panel spec (#381) (#398) Co-authored-by: UX/UI Developer 1 <ux-dev-1@roboco.tech> * [1fb5b5cb] Project picker, re-render button, and composition preview panel (#397) (#402) * [1fb5b5cb] feat(video): project picker, re-render button, and composition preview panel * [1fb5b5cb] docs(video): add comprehensive guide for project picker, re-render button, and composition preview panel --------- Co-authored-by: Frontend Developer 1 <fe-dev-1@roboco.tech> Co-authored-by: Frontend Documenter <fe-doc@roboco.tech> * [a512f364] Add video_engine_enabled to ProjectSummaryResponse (#412) (#414) * [a512f364] feat(api): surface video_engine_enabled on ProjectSummaryResponse * [a512f364] docs(api): document video_engine_enabled on ProjectSummaryResponse --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [03607ab9] Fix re-render control gating/placement and project picker filter (#434) * [f2f3e89f] Fix RerenderControl gating/placement across queue and strip views (#431) * [f2f3e89f] feat(video): widen RerenderControl gating and share it across queue/strip views Extracts RerenderControl into a shared panel/src/components/dashboard/ video-rerender-control.tsx component, widens its gate from render_status === 'failed' to source_task_id + composition_id present (matching what the backend rerender endpoint actually requires), adds a confirm dialog before firing the mutation, and wires the same component into video-pipeline-strip.tsx for still-in-flight rendering/render_failed rows. * [f2f3e89f] docs(video): enhance RerenderControl JSDoc with gating logic and usage examples Add comprehensive JSDoc to the RerenderControl component covering its purpose, gating logic (render for any source_task_id + composition_id, regardless of render_status), three visual button states (idle/loading/ error), confirm-dialog guard behavior, and usage examples for both video-post-queue.tsx and video-pipeline-strip.tsx contexts. Explains why the backend's rerender endpoint doesn't require a failed render and how the component prevents accidental re-renders. --------- Co-authored-by: Frontend Developer 1 <fe-dev-1@roboco.tech> Co-authored-by: Frontend Documenter <fe-doc@roboco.tech> * [404d8ed3] Filter project picker to video-engine-enabled projects (#432) * [404d8ed3] feat(panel): filter video-request project picker to opted-in projects Add video_engine_enabled to the client ProjectSummary type, give ProjectSelector a videoEngineOnly filter prop, default RequestVideoDialog's picker to the current video-enabled project with a friendly empty-state when none exist, and cover the filter with a new project-selector test. * [404d8ed3] docs(panel): add ProjectSelector component API reference with videoEngineOnly filter Document the reusable ProjectSelector component with its props, filtering behavior, and new videoEngineOnly filter for video-engine-enabled projects. Follows the existing component documentation pattern from page-refresh-provider. --------- Co-authored-by: Frontend Developer 2 <fe-dev-2@roboco.tech> Co-authored-by: Frontend Documenter <fe-doc@roboco.tech> * [519a4088] fix(panel): import missing RerenderControl in video-post-queue and correct stale doc (#436) Co-authored-by: Frontend Developer 1 <fe-dev-1@roboco.tech> --------- Co-authored-by: Frontend Developer 1 <fe-dev-1@roboco.tech> Co-authored-by: Frontend Documenter <fe-doc@roboco.tech> Co-authored-by: Frontend Developer 2 <fe-dev-2@roboco.tech> * [8e912c3e] Reflow hard-wrapped video UX design doc to pass quality gate (#439) * [ccfe2015] docs(ux_ui): reflow video request composition-controls spec to one line per paragraph (#438) Co-authored-by: UX/UI Developer 1 <ux-dev-1@roboco.tech> * [99c3ed9c] docs(backend): reflow hard-wrapped prose in video-engine-endpoints.md and video-project-scoping.md (#442) Co-authored-by: UX/UI Developer 1 <ux-dev-1@roboco.tech> * [c2e98fc0] docs(backend): strip stray trailing whitespace in video-engine-endpoints.md fence (#451) Co-authored-by: UX/UI Developer 1 <ux-dev-1@roboco.tech> * [9c7bc11a] Reflow all 3 hard-wrapped docs on this branch and verify quality gate (#457) * [9c7bc11a] test(scripts): guard reflow_md.py --check wiring into make quality * [9c7bc11a] docs(standards): document markdown reflow quality gate workflow and verification Added comprehensive guide explaining the one-logical-unit-per-line markdown prose standard, how the reflow check integrates into make quality, the three reflowed files (video-engine-endpoints.md, video-project-scoping.md, composition-controls spec), and the regression test added to ensure wiring stability. This task verifies all three ACs are satisfied: reflow_md.py --check exits 0, make quality passes (non-DB portions), and the three files are whitespace-only reflowed. --------- Co-authored-by: UX/UI Developer 1 <ux-dev-1@roboco.tech> Co-authored-by: UX/UI Documenter <ux-doc@roboco.tech> --------- Co-authored-by: UX/UI Developer 1 <ux-dev-1@roboco.tech> Co-authored-by: UX/UI Documenter <ux-doc@roboco.tech> * [002f0cdd] docs(rag): document reflow-check zero-diff troubleshooting path (#459) (#460) Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> * [e4ed92d6] fix rerender missing-task test — assert the empty queue it creates The test never seeds; the trailing assertion expected a phantom video post. Broken since the branch's first commit but never executed — every earlier CI run short-circuited at a pre-pytest gate step. --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> Co-authored-by: UX/UI Developer 1 <ux-dev-1@roboco.tech> Co-authored-by: Frontend Developer 1 <fe-dev-1@roboco.tech> Co-authored-by: Frontend Documenter <fe-doc@roboco.tech> Co-authored-by: Frontend Developer 2 <fe-dev-2@roboco.tech> Co-authored-by: UX/UI Documenter <ux-doc@roboco.tech> Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
4d52f6ff59 |
[1f6a06a2] PR-review gate: verify ACs literally and require green CI before pr_pass (#428)
* [a1bde3b9] Add CI-status guard to pr_pass + update pr_reviewer prompt (#417) (#420) * [a1bde3b9] feat(gateway): CI-status guard on pr_pass + reviewer prompt update * [a1bde3b9] docs(pr-gate-review, worksession-git): document CI-status guard on pr_pass Updated two architecture documentation files to reflect the new CI-status guard: **pr-gate-review.md:** - Documented _ci_status_guard method: blocks pr_pass on failing/pending/unscheduled/error CI with reviewer-aware pr_fail remediation - Documented _resolve_ci_status: best-effort GitHub check-runs lookup with fail-open behavior - Updated _pr_pass_blocked description: now returns (rejection_envelope, ci_note) tuple - Updated _record_gate_verdict_for/verdict to note ci_status field stamping on pr_pass - Added ci_note parameter documentation for evidence tracking when no CI is configured - Updated Logical Tree to show new methods - Added Config Flags note: CI guard is always armed, fails open on config gaps - Added two regression risks: check-runs-only limitation, fail-open design **worksession-git.md:** - Documented GitService.get_pr_ci_status(project_slug, pr_number): CI status lookup with state classification - Documented supporting methods: _ci_status_prereqs, _fetch_check_runs, _classify_check_runs, _classify_zero_check_runs - Each method notes its fail-open behavior and configuration gap handling --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [e8f275d7] test(gateway): lock the 7-AC-to-test map + assert pr_reviewer prompt content (#425) (#426) Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> * [24b4237e] Fix reflow-check, CI-status classification, and noqa suppression (#440) (#443) * [24b4237e] fix(gateway): classify unreachable/nonexistent CI-status repo as no_ci_configured, remove test noqa, reflow pr_reviewer.md Split GitService.get_pr_ci_status's PR-head-sha lookup into a dedicated helper so a config gap (missing project/git_url/token) or an unreachable/ nonexistent repo/PR (network error or 404) classifies as no_ci_configured (pr_pass passes through and stamps the evidence note) while a genuine GitHub API failure on a real, reachable repo (any other non-2xx, or an unparseable body) stays the fail-closed error state. Replaced the `# noqa: PLR2004` in test_git_pr_ci_status.py with a named HTTP-status range constant, updated the config-gap tests to assert the new classification, and added tests for the unreachable-repo and real-repo- API-error branches. Reflowed agents/prompts/roles/pr_reviewer.md's one hard-wrapped continuation line so it passes make reflow-check. * [24b4237e] docs(gateway): update pr-gate-review.md for CI-status classification refactor Updated the internal architectural map to reflect the new CI-status classification scheme introduced in PR #440. Configuration gaps (missing project/git_url/token) and unreachable/nonexistent repos (404 or network error) now explicitly classify as no_ci_configured and pass through with evidence stamps. Genuine GitHub API failures on reachable repos classify as error and stay fail-closed (retryable). - Clarified _ci_status_guard behavior: config gaps/unreachable repos pass through with distinct classification; only real API failures stay fail-closed - Updated Config Flags section to describe the new three-way classification - Updated Regression Risks section to document the new explicit classification scheme - Noted that _resolve_ci_status now wraps git.get_pr_ci_status and interprets its result dict --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> * [1f6a06a2] round-3 fixes: pr_gate back to xenon rank A; 404 means no CI, not error Eight extracted helpers bring the module average from B(5.05) to A(4.04) with every external contract untouched (170 gate tests byte-identical). The CI-status guard now classifies a 404 on the check-runs or workflows endpoints as no_ci_configured (pass-through with evidence note) — a repo without Actions is not a transport failure — reserving the fail-closed error state for network/5xx/auth failures, with pinning tests for all four shapes. The e2e fake-GitHub router gains check-runs and workflows routes so the scripted lifecycle exercises the guard's green-CI success branch end to end. * [1f6a06a2] merge master; align gate-diff-base tests with the tuple contract The merged tree is the first integration of the CI-status guard with the preferred-parent diff-base guard: _pr_pass_blocked now returns (rejection, ci_note), so the diff-base tests unpack it instead of asserting on a bare result. Both guards verified live in the merged pr_gate (preferred_parent threading and _ci_status_guard present). --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Backend Documenter <be-doc@roboco.tech> Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
15a3e87a2f |
feat(vault): Obsidian vault V1 — projection core, Auditor narrative, input loop (#458)
* fix(gateway): gate review diffs against the task's real parent branch (#444) The in-path PR-review gate's evidence diff (claim_gate_review) and the pr_pass conventions guard derived their diff base via parent_branch_for string surgery, which reuses the child branch's own team segment — wrong for every cross-team hop (a frontend child of a main_pm root derives a ref that never existed) and silently falls back to the repo default branch, so the reviewer judged the entire inherited base-branch content as the task's own work and failed acceptance criteria the task never touched. Bounced a live goals-tab fix three times, unfixable by branch surgery. The gate now resolves the base via resolve_parent_branch (the parent task's recorded branch_name, cross-team correct) and threads it as a new preferred_parent override through git.diff / list_changed_files / conventions_check_for_task — consulted only when no explicit base is given, so the pinned literal-base contract (base="HEAD~1") and every other diff caller (QA, doc, content) are byte-identical. Parent lookup fails open (derived-base fallback) like the other resolve_parent_branch call sites, and is skipped entirely while the conventions flag is off. Also excludes .uv-cache/ and .claude/ (agent worktrees, private uv cache) from the markdown prose scanner — both are repo-local tool dirs whose vendored/generated files tripped make reflow-check. Co-authored-by: Renn F <rennf93@users.noreply.github.com> * feat(vault): Obsidian vault V1 — projection core + input loop The vault is a rebuildable projection of the DB (never a source of truth), default-off behind ROBOCO_OBSIDIAN_VAULT_ENABLED + ROBOCO_VAULT_PATH. Projection core: VaultWriter materializes tasks/journals/A2A/agents as wikilinked markdown (id-suffixed stable filenames, alias-based links so renames never break, is_private journals excluded like the RAG corpus); event seams materialize on journal write and A2A send and touch task-note frontmatter at the status-transition chokepoint — all best-effort, a vault failure never blocks a verb. Shipped .obsidian config (Dataview, Kanban, team/status graph groups) + _meta dashboards; python -m roboco.vault rebuild/relocate (rebuild preserves the narrative section). Auditor narrative duty: curate_vault content verb (auditor-only, playbook-curation pattern) spawned by a dedicated root-completion hook with its own cooldown — fully separate from _dispatch_audit_work, whose scheduled-sweep/alert-producer revival belongs to the queued fleet task. Input loop: VaultIntakeEngine (ROBOCO_VAULT_INTAKE_*) watches the intake folder for #roboco-tagged notes and materializes each as ONE held draft (confirmed_by_human=False, Secretary-owned, source=vault_note, excluded by the dispatchers via _is_held_ceo_source) with local-model extraction and a deterministic fallback; vault_seen_notes ledger (migration 070) keyed on path+content-hash (the CEO-feedback callout is stripped before hashing so the engine's own append never self-triggers); per-cycle and open-draft caps. Nothing auto-starts. * fix(vault): integrate with master — re-chain migration 070 onto 069, mypy-clean tests The vault branch was cut from slave before the sequence-gate promotion, so migration 070 chained from 068 while master already carried 069 — two heads on merge. Master is now merged in and 070 revises 069. Vault test files also get mypy-clean mock idioms (monkeypatch.setattr over method assignment; await_args narrowed before access). * fix(scripts): dedupe SKIP_DIRS again after the master merge Master still carries the twin-merge duplicate (its dedupe hotfix is an unmerged PR); the merge re-imported it here. * fix(vault): reflow hard-wrapped prose in the vault asset templates * chore(config): exclude .uv-cache and .claude from deptry's scan scope Same repo-local tool dirs the prose scanner skips; the standalone deptry target walks the repo root and drowned in the cache's unpacked wheels. * fix(vault): board-review activation path for vault drafts + relocate graft The input loop's held-artifact posture was a dead end: vault_note drafts were unconditionally held by _is_held_ceo_source, owned by the verbless Secretary, hidden from the panel's approval surfaces by team, and the open-drafts cap counted them forever — the engine self-bricked after ten notes. Vault drafts now ride the intake board-review path instead: a tagged note becomes a PENDING Product-Owner-assigned Board draft (the exact confirm_live_draft board shape), the board reviews it, and only the CEO's approve_and_start makes it deliverable — never-auto-starts now rests on the board gate, proven by tests against the real dispatchers. The cap counts only drafts still awaiting the CEO (team==BOARD, non-terminal), so approval and cancellation both free it. relocate into an existing personal vault now grafts old_root/RoboCo as a direct child (refusing loudly if RoboCo/ already exists there) and adds only absent .obsidian/_meta files — a personal vault's config is never clobbered. An absent destination keeps the whole-tree move. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
8f3f4236c0 |
feat(tasks): sequence is the bar — strict sibling ordering at the claim chokepoint (#452)
* feat(tasks): enforce sibling sequence order at the claim chokepoint A task with a parent and effective sequence N (COALESCE(sequence, 0)) can no longer be claimed while any sibling with a strictly lower effective sequence is non-terminal — assignee-blind, independent of and stricter than dependency_ids, enforced in _validate_claim_preconditions so both claim paths (gateway verbs and the dispatcher's raw REST claim) cross it. Ties run parallel; cancelled siblings never block; sequence 0 and parentless tasks are unaffected. Live failure this guards: a PM delegated revision subtasks sequenced 0..3 with no dependency edges and seq 2 started alongside seq 0 — sequence was advisory-only. set_sequence's contract updated accordingly. New e2e smoke case drives the refusal and the post-completion claim through the real gateway. * chore(scripts): skip .uv-cache and .claude in the prose scanner Repo-local tool dirs (private uv cache, agent worktrees) carry vendored and generated markdown that tripped make reflow-check. * fix(tasks): wave-derived delegation sequences + claim-gate hardening Three fixes from the adversarial review of the sequence claim gate: Delegation no longer stamps a raw per-sibling ordinal (deterministic merge-order bookkeeping) as sequence — under the strict gate that serialized ALL delegated work, including fully independent cross-dev and cross-cell siblings. Sequences are now wave-derived post-wiring (stamp_wave_sequence: 1 + max same-parent dependency sequence, 0 when independent), so independent siblings tie and run parallel while colliding/ordered work ascends. The cross-cell UX wiring restamps instead of writing relative ux+1 values (a relative write could invert a collision-derived stamp), and the dispatch merge/lane barriers gain a created_at tiebreak for wave-tied siblings so shared-branch merge order stays deterministic. PM-authored sequences are never rewritten. The guard now also fires on reclaims from needs_revision (a lower- sequence sibling delegated after the first claim was invisible), and tasks.parent_task_id gains an index (migration 069) — the guard's sibling probe ran as a Seq Scan on the hottest verb. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
7ff70ab5e2 |
fix(gateway): gate review diffs against the task's real parent branch (#444) (#454)
The in-path PR-review gate's evidence diff (claim_gate_review) and the pr_pass conventions guard derived their diff base via parent_branch_for string surgery, which reuses the child branch's own team segment — wrong for every cross-team hop (a frontend child of a main_pm root derives a ref that never existed) and silently falls back to the repo default branch, so the reviewer judged the entire inherited base-branch content as the task's own work and failed acceptance criteria the task never touched. Bounced a live goals-tab fix three times, unfixable by branch surgery. The gate now resolves the base via resolve_parent_branch (the parent task's recorded branch_name, cross-team correct) and threads it as a new preferred_parent override through git.diff / list_changed_files / conventions_check_for_task — consulted only when no explicit base is given, so the pinned literal-base contract (base="HEAD~1") and every other diff caller (QA, doc, content) are byte-identical. Parent lookup fails open (derived-base fallback) like the other resolve_parent_branch call sites, and is skipped entirely while the conventions flag is off. Also excludes .uv-cache/ and .claude/ (agent worktrees, private uv cache) from the markdown prose scanner — both are repo-local tool dirs whose vendored/generated files tripped make reflow-check. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
bba20a3917 |
feat(prompter): board-review → redraft loop for MegaTask batches (#411)
Batch parity with the single-draft keep-alive redraft loop. A first board-route confirm-batch parks the intake session against the umbrella (instead of the unconditional reap), so the existing board-completion injection reaches the still-live chat — now with a batch-aware brief (compose_batch_redraft_message: live root-subtask snapshots + board notes + a one-propose_batch re-proposal instruction). The re-confirm carries BatchConfirmRequest.task_id and routes to the new PrompterService.update_live_batch: in-place umbrella + root-subtask update (positional patch of live children, cancel+recreate on scope change, create/cancel on count change, dependency edges rewired to the fresh wave plan) gated by the same _validate_batch_scope as create. Readers use the CANCELLED-excluding get_live_subtasks view so multi-round redrafts survive earlier cancels. Cold path: re-interview now handles a branchless umbrella by recovering its multi-repo scope from live children (distinct_projects_for_batch) and returning project_ids — fixes the live 400 behind the task-detail redraft button on umbrellas. Panel: confirmBatch board branch keeps the chat open, threads batchRedraftTaskIdRef (persisted) into the re-confirm, treats a redraft re-confirm as terminal on both routes, and surfaces the server's real validation message on confirm failure. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
3674a1e002 |
fix(megatask): wire cross-cell sequencing for batch root-subtasks (#391)
Within a MegaTask root-subtask, the per-cell tasks got sequence numbers but zero dependency edges, so they ran fully in parallel (UX finished after backend started, frontend self-blocked) — divergent branches, duplicated/wasted work. The cross-cell wiring (_wire_ux_frontend_dependency: FE/BE cells depend on the UX cell, bidirectional, propagated to dev subtasks via inherit_unmet_dependencies) already exists, but it bails unless the parent has a product_id. A MegaTask root-subtask has no product_id — it targets its cells via cell_projects — so the wiring silently no-op'd for every MegaTask root (confirmed on the live video root: product_id=None, three cells, all with empty dependency_ids). Broaden the guard to fire on product_id OR is_batch_root_subtask(batch_id, parent_task_id) (scalar fields; cell_projects is a lazy relationship). The same tested wiring now holds MegaTask cells in order like a product fan-out. Adds test_megatask_root_wires_cross_cell_ux_dependency. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
91f9642f27 |
fix(megatask): guardrail the wave sequence at the claim chokepoint (#382)
The Main PM claimed every MegaTask wave at once, ignoring the collision-ordered
dependencies. The sequencing data was correct (analyzer wired proper waves), but
enforcement was only half-wired: the unmet-dependency guard lives on the gateway
claim verbs (i_will_plan -> _run_claim_guards), while the orchestrator dispatches
coordination roots itself — _dispatch_pm_work fetches pending with no dependency
filter and _claim_task_for_agent system-claims via the raw POST /tasks/{id}/claim
route -> TaskService.claim, which had no dependency check. So the orchestrator
claimed every pending root-subtask for the Main PM regardless of wave.
Enforce the sequence at the claim chokepoint: _validate_claim_preconditions now
refuses to claim a PENDING task while any depends_on task is non-terminal
(extracted into _claim_blocked_by_dependencies for the complexity budget). This
guardrails every claim path — the gateway verbs (redundant) and the orchestrator
raw dispatch claim (the hole). Scoped to a PENDING start-of-work claim so a
mid-lifecycle QA/doc claim is unaffected; dependencies are monotonic so each wave
claims normally once the prior one completes.
Adds test_claim_pending_with_unmet_dependency_returns_none (blocked with an
unfinished dependency; claimable once it completes).
Co-authored-by: Renn F <rennf93@users.noreply.github.com>
|
||
|
|
f601e32788 |
fix(prompter): batch confirm ignores a vestigial top-level project slug (#380)
* fix(prompter): batch confirm ignores a vestigial top-level project slug The MegaTask confirm-batch 400'd with "Invalid project_id UUID: roboco-api". The intake agent authors each draft's project as the repo slug it read, and the panel only nulls the top-level project_id when the CEO toggles that draft's picker — so an untouched draft carried the slug through to create_task_from_draft, whose eager _resolve_uuid_field(project_id) raised on the non-UUID and rejected the whole batch (guard-release from the prior fix surfaced it as a clean 400 instead of a wedged 500). A batch root-subtask targets its repos via the the_work per-cell map (the panel fills those with real project UUIDs); the top-level project_id/product_id is vestigial for it. Strip it from a sub-draft when its cell map carries the real target — a legacy no-the_work draft keeps its panel-filled top-level UUID, and scope validation (which already runs off the cell map) is unchanged. Adds a repro test with two cell-map drafts that also carry leftover top-level slugs (roboco-api / roboco-panel); they now confirm instead of 400-ing. * fix(tests): conventions PR integration test honors the #375 workspace-scope guard #375 added a containment guard to open_conventions_pr (workspace_path must sit under {workspaces_root}/{slug}); the unit test was updated but this integration test still seeded a bare tmp_path/repo, so open_conventions_pr returned None and test_open_conventions_pr_commits_locally_without_remote failed on master. Anchor workspaces_root at the test dir and place the repo under the project's slug. * refactor(prompter): extract batch sub-draft sanitize (xenon rank B) The inline vestigial-target strip pushed _build_confirm_batch to cyclomatic rank C (over the --max-absolute B gate). Move the assigned_to + top-level project/product stripping into a pure _batch_subtask_draft helper; behavior is unchanged, _build_confirm_batch drops back under the limit. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
2297d448f3 |
fix(prompter): keep a watched intake chat alive (idle-reap counted reading as idle) (#379)
* fix(prompter): keep a watched intake chat alive (idle-reap counted reading as idle) Intake chats "dropped after a while" — the panel showed "Live connection lost". The idle reaper retires an interactive session whose last_activity is older than interactive_idle_reap_seconds (30m default), but last_activity was bumped only by an agent event or a human turn. An open SSE stream — the human reading a proposed draft / MegaTask spec without typing — bumped nothing, so a chat under active review was reaped mid-read, closing the stream (the SSE transport error the panel reports as "Live connection lost"). stream() now runs a keepalive task that refreshes last_activity every 60s while the stream is connected, so an open, actively-watched chat counts as alive; when the tab closes the generator ends, the keepalive is cancelled, and a genuinely abandoned chat still reaps after the threshold. The keepalive runs beside an un-cancelled queue.get() so no live token or the close sentinel can be dropped. * fix(tests): conventions PR integration test honors the #375 workspace-scope guard #375 added a containment guard to open_conventions_pr (workspace_path must sit under {workspaces_root}/{slug}); the unit test was updated but this integration test still seeded a bare tmp_path/repo, so open_conventions_pr returned None and test_open_conventions_pr_commits_locally_without_remote failed on master. Anchor workspaces_root at the test dir and place the repo under the project's slug. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
3787d15524 |
fix(agents): block subagent spawning at the Claude Code level (disallow Task) (#377)
The fleet-wide subagent ban was implemented as an allowlist omission, but Task is a default-permitted Claude Code built-in — an allowlist auto-approves, it does not restrict. Under permission_mode="dontAsk" (intake/secretary SDK) and defaultMode="bypassPermissions" (fleet), Task ran regardless and can_use_tool was never invoked for it, so every Claude-path agent could still spawn subagents despite allows_subagent=False. Only the grok path blocked it. Explicitly disallow the subagent tool at every Claude-path spawn point: disallowed_tools=["Task"] on the intake and secretary SDK drivers, and "Task" in the fleet settings.json base_deny (an explicit deny applies even under bypassPermissions). This mirrors the grok path's --disallowed-tools Agent. Pins the ban in test_cc_lockdown.py (fleet settings deny Task) and a new test_sdk_driver_subagent_ban.py (intake + secretary options disallow Task). Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
92410d47cf |
fix(prompter): MegaTask review card scrolls + confirm-batch guard releases on failure (#376)
The MegaTask review card was a static sibling of the scrollable chat list with
no height bound, so a tall batch overflowed the clipped container and stranded
the launch buttons off-screen with no scrollbar. It now owns the scroll area
(min-h-0 flex-1 overflow-y-auto), matching the pattern ChatMessages already uses.
confirm_live_batch's Redis idempotency guard (1h TTL) was acquired before the
build but only released via a success sidecar, so a build failure wedged the
session: every retry hit ServiceError("already in progress") -> HTTP 500 for an
hour. The build now releases the guard on any failure before the sidecar write,
so a retry re-attempts (and surfaces the real error) instead of being locked out.
Extracted _build_confirm_batch to keep the try/except thin.
Adds DB-backed tests for the panel the_work[].project_id shape, multi-cell
root-subtasks, dense same-repo collisions (both routes), and guard release.
Co-authored-by: Renn F <rennf93@users.noreply.github.com>
|
||
|
|
7be725cc13 |
fix(security): disposition all 104 code-scanning + dependabot alerts (#375)
Fix the 4 real CodeQL path-injection alerts (open_conventions_pr trusted the API-settable project.workspace_path with no containment) plus defense-in-depth segment validation at the get_workspace_path chokepoint. Bump next 16.1.1->16.1.7 and transitive lockfile deps to clear 24 Dependabot alerts. Close the intake subagent-ban gap: the Claude intake driver still carried the Task tool and the prompter prompt told it to fan out research subagents, contradicting the fleet-wide ban. The remaining 47 CodeQL + 29 Dependabot alerts are dismissed on GitHub with per-alert justifications (guard patterns CodeQL can't model across call hops; next 16.2.x blocked by the verified tab-hostage router regression). Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
0450ec9e89 |
feat(x-engine): smart spotlight cadence — daily when there's news, quiet when there isn't (#374)
CEO verdict on the blind 3-day timer: 'It should default to 1 day and be more like... smart.' Now: interval defaults to 1 day; the cycle skips (with logged reasons) while a spotlight draft is still awaiting the CEO, and stretches to 3x the interval when nothing has shipped (CHANGELOG sections via the read clone) since the last spotlight activity — where activity is a materialized draft's seen_at or a completed exploration's updated_at, deliberately excluding the stale- cycle janitor's cancels. The HoM gains an explicit skip exit (propose_feature_spotlight skip=true + reason: completes the exploration, no draft, no seen-slug, still counts as activity), and its spawn prompt now carries the seen ledger WITH dates, what shipped since the last spotlight, and recently rejected drafts with the CEO's reasons — fresh-but-unspotlighted first. Fail-open on changelog read errors so a signal outage never starves the engine. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
Renn F
|
1c799ea139 |
fix(tests): grok session test pins the fleet-wide subagent ban
The last test asserting the prompter kept the Agent tool — it pinned the allowance through the disallowed-tools string, which the #372 test sweep (grepping the flag names) missed. |
||
|
Renn F
|
e1a8734daa |
fix(tests): hoist ROLE_CONFIGS import in the subagent invariant test
PLC0415 red on master's quality gate — the module already imports ROLE_CONFIGS at top level; the function-level re-import from #372's invariant test is deleted. |
||
|
|
7bc42ec114 |
fix(runtime): fleet-wide subagent ban — no role fans out, not one (#372)
CEO directive 2026-07-09: every allows_subagent in role_config flips to False (was True for cell_pm, main_pm, product_owner, head_marketing, prompter, secretary); the grok path's drifted _SUBAGENT_ALLOWED_ROLES allowlist empties to match. The spawn manifest already consumes the flag, so Claude-path agents lose the Agent tool and grok-path agents get it in disallowed-tools by construction. New invariant test iterates every role config so a single role can't quietly regain it. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
Renn F
|
93f63d5fe9 |
fix(gateway): allow sync_branch onto a legitimately-master base
A standalone task (video/CI-watch/dep-update — no parent) and a child of a branchless coordination parent merge into the project default branch by design, but the protected-base guard refused every master/main base, hard-wedging their rebases into block/PM/respawn churn (hit live on the v0.19.0 video task). The rebase force-pushes only the task branch (with lease) and cannot write to the base, so the guard now refuses master/main only when it is mis-resolved: a branch-bearing parent exists or the parent row is missing/corrupt. The '-'-prefixed injection guard stays unconditional. |
||
|
|
18998c4a42 |
Video pipeline fixes: visibility strip, rich briefs, spotlight timing + fps (#369)
* feat(video): pipeline visibility — strip, state-aware queue, render-error capture Task 1 of the 2026-07-09 video-pipeline review. New CEO-gated GET /video/pipeline lists every in-flight video item (authoring statuses, rendering attempt n/max, terminal failures with the error — now stamped onto the video_draft marker instead of dying as a log line). source_task_id exposed on both video schemas. Panel: pipeline strip on the Social page, state-aware queue empty copy, title/script on queue rows, missing cuts disabled instead of a blank player, notifications deep-link related_task_id. MAX_VIDEO_RENDER_ATTEMPTS moved to the markers policy layer (single source of truth). * feat(video): rich authoring briefs — changelog section, brand voice, kit pointer Task 2 of the 2026-07-09 video-pipeline review. The release brief is now a structured block (full CHANGELOG section capped at 4000 chars + highlights) instead of one LLM-compressed sentence; brand_voice and a motion/kit design-bar pointer are appended centrally in open_video_task so release, spotlight, and on-demand paths all inherit them. suggested_input_props seeded on the video_draft marker; third acceptance criterion pins the design bar; propose_video docstring points at the kit. * fix(video): spotlight video drafts on CEO approval, renderer honors data-fps Task 4 of the 2026-07-09 video-pipeline review. The companion-video hook moves from propose_feature_spotlight (HoM authoring time) to XPostService approve's posted-success branch for x_feature drafts, mirroring the release-publish seam — a rejected spotlight no longer burns a ux-dev cycle; wants_video/video_script ride the x_feature_ref marker. Best-effort: a video-engine failure never breaks the post. render.js reads data-fps from the composition HTML (clamped 24-60, fallback 30) instead of hardcoding 30; parseFps covered by node --test. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
ec5323917e |
[d8a2999b] CI-watch: fix the CI regression on roboco-api (#364)
* [ffcc7317] fix(tests): tolerate tagless checkouts in release-readiness smoke test (#362) (#363) Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> * docs: reflow panel/UPGRADE.md — no hard-wrapped prose --------- Co-authored-by: Backend Developer 1 <be-dev-1@roboco.tech> Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
47c927c598 |
feat(gateway): root-owned acceptance criteria via declare_coverage (#357)
The coverage gates had no vocabulary for criteria only the root itself can satisfy (the supersede PR from feature/main_pm/*, closing the contributor's PR): once a Main PM declared coverage for the legitimate cell criteria, the idle gate demanded a cell for the impossible ones too, so they got pushed into a cell task and the cell PM (correctly) escalated. declare_coverage now accepts the PM's own task: self-declared criteria count as claimed for the idle gate and satisfied for the roll-up (the roll-up actor is their owner by construction), surface as claimed_by=root in the briefing, and both PM prompts say to never hand a cell a criterion it cannot satisfy inside its own cell. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
199cc5d2bc |
fix(api): run the RAG reconcile in the background, never on the bind path (#346)
The lifespan awaited the reconcile, and the journal/learning backfill made it expensive: up to 200 entries each needing an Ollama embedding behind a busy Ollama held the API bind down for 30+ minutes on the NAS (observed: ~6 embeds/min). uvicorn only binds after the lifespan completes, so the whole stack 502'd while a best-effort maintenance pass ran. The reconcile is now a background task scheduled at the end of startup (crash-logged via done-callback, cancelled at shutdown); the backfill still converges across boots under its per-boot cap. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
f0b6390189 |
feat: Social page — aggregated post queues + X/video history (#345)
* feat(api): x/video post history endpoints Approved or rejected drafts vanished from both queues permanently -- the listers exclude terminal statuses and no history surface existed, so a posted tweet or video was only findable in the raw task list. GET /x/posts/history and GET /video/posts/history (CEO-gated, bounded) return acted-on drafts newest-first with the posted platform ids and reject reasons from the draft markers. Route tests assert by identity, not emptiness: approve/reject commits the whole session, so prior tests' rows legitimately persist in the shared test DB. * feat(panel): Social page aggregating post queues and history New dashboard page composing the X and video post queues with one unified history section beneath them -- both platforms interleaved newest-first, kind and outcome badges, posted X ids linking to the live tweet, reject reasons shown. The command center's two full queue cards become a compact pending-counts card linking to the page, so the queues have one home instead of duplicated surfaces. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
5886336259 |
fix(db): latch init_db per database and time-bound the alembic runner (#342)
Bootstrap and the API lifespan both ran init_db in one process seconds apart; the second call re-entered the alembic-in-thread machinery (nested asyncio.run + NullPool engine + greenlet bridge in a reused worker thread) for zero benefit and hung two consecutive NAS boots there, blocking the API bind forever with zero SQL activity. init_db now latches per database URL (drop_db resets it; a different DB always runs fully), and the alembic worker is bounded at 300s -- a wedged thread fails startup loudly with a pinpointed error so the container restarts into a clean retry instead of hanging silently. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
Renn F
|
5e7c498d00 |
fix(runtime): auto-submit is unconditional; refusals brief the fallback PM
The PR-gate turn cut (#295) already auto-submitted assembled tasks, but its refusals fell back to a PM spawn silently -- in production the PM turns the cut was meant to remove kept happening with no visible cause (live case: an AC-coverage refusal). The flag is gone (the fallback is the safety net), the umbrella/branchless exclusion uses the canonical batch predicates, and every refusal reason now rides into the spawned PM's prompt so the fallback starts informed. |
||
|
|
f48d088c08 |
fix(gateway): working exits for wedged agents + declare_coverage roll-up unblock (#341)
A live task burned 5+ hours because every exit was locked. unclaim now works from verifying and needs_revision (service guard + lifecycle edge); the circuit breaker and the i_am_done push-failure remediate name the working chain ending in unclaim(); sync_branch(stash=true) clears the DIRTY_WORKSPACE dead-end (pop-conflict preserves the stash); blocking a task QA already owns now says to idle instead of listing states; the orchestrator auto-block logs real errors and skips states where blocking is meaningless instead of force-blocking them. declare_coverage (cell/main PM) retroactively stamps parent-AC refs on a child that implements them -- closing the roll-up deadlock where the declaring child was cancelled and its re-delegated replacement completed the work uncredited. Cancelling a ref-declaring child now warns and surfaces the orphaned criteria. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
47d78f50ee |
feat(sandbox): on-demand provisioning via request_sandbox verb (#338)
* feat(sandbox): on-demand request_sandbox verb replaces eager provisioning Sandboxes were provisioned at every agent spawn for opted-in projects, so every role paid the sidecar spin-up and a provisioning failure refused the spawn. Provisioning now happens when an agent asks: the request_sandbox do-verb (dev + QA) reaches the orchestrator through ContentActionsDeps, ensure_sandbox provisions idempotently with an in-memory per-agent cache (evicted at teardown and janitor sweep), and creds return in the envelope payload including ready-to-export ROBOCO_TEST_* values. Spawn now only injects a marker env naming the available services plus a briefing line; sandbox failures can no longer refuse a spawn. Teardown lifecycle unchanged. * feat(sandbox): harden request_sandbox + Phase 3 wiring proof and docs Hardening from adversarial review: ensure_sandbox now provisions the project's full opted-in set on first request (a later superset can never tear down a live sandbox mid-use), serializes per-agent behind an asyncio lock (a client timeout-retry no longer races its own in-flight provision), and verifies container liveness on every cache hit (a dead sandbox evicts and re-provisions instead of serving dead creds). MCP client budget 720->1080s for the full-set cold case. Phase 3: e2e smoke wiring test (manifest grants + guard-chain envelopes over the real API), sandbox-db/tools/map docs and CLAUDE.md rewritten for on-demand. * feat(sandbox): release sandboxes when the agent's work ends CEO directive: sidecars must not dangle once the agent is done. The six work-ending verbs (i_am_done, unclaim, i_am_idle, pass_review, fail_review, i_documented) now release the caller's sandbox best-effort on their success path via release_sandbox (lock + teardown + cache evict; a no-sandbox agent costs a dict lookup). Container removal and the janitor remain the backstop; a re-request provisions fresh. * test(sandbox): monkeypatch the release hook instead of method assignment mypy method-assign rejected the direct AsyncMock assignments; the prior static gate ran before this test file landed. * test(sandbox): guard envelope evidence for mypy in verb tests * chore(prompts): regenerate verb tables for request_sandbox * chore: resolve merge with master (breadcrumbs + statement budget) --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
9e4025b822 |
fix(git): scope post-op ownership repair to what the op could change (#337)
Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
4fb0059556 |
fix: journals/learnings never reached the RAG corpus + git-readonly slug 404s (#339)
* fix(rag): per-index chunk floors — journals and learnings were never indexed The global 200-char garbage floor (sized for code/doc chunks) discarded every templated journal note and most distilled org-memory lessons, silently: ingest returned success with zero chunks, so agent journals and learnings were never retrievable via RAG. IndexConfig now carries a per-type min_chunk_length (journals 40, learnings 80, others unchanged). * fix(mcp): git-readonly tools default project_slug from the container env Agents 404ed /api/git/status with 'Project not found: roboco' — the tools made the LLM supply the slug and six doc examples taught a slug that matches no registered project. The tools now fall back to the ROBOCO_PROJECT_SLUG the orchestrator already injects, and the stale examples are corrected. * feat(rag): startup backfill re-ingests zero-chunk journals and learnings Before the per-index chunk-floor fix, ingest() returned success with chunk_count=0 for undersized content: every historical journal entry and distilled learning below the (then-global) 200-char floor was durably recorded in journal_entries but silently never got a chunks_journals / chunks_learnings row, and no exception meant the existing dead-letter (rag_index_failures) never saw it either. Extends the startup reconcile (roboco/api/app.py _reconcile_rag_indexes) with a new pass: backfill_unindexed_journals (roboco/services/ rag_index_failures.py) queries journal_entries for rows missing from each vector table and re-ingests them through the same live code paths (_reindex_journal_entry / record_learning). Journals and learnings are backfilled independently since a LEARNING entry can clear the (lower) JOURNALS floor while still failing the (higher) LEARNINGS floor — a learning's doc_source is a content hash, not the entry id, so presence there is checked by hashing each candidate the same way LearningsIndexPlugin.record_learning does and batch-querying chunks_learnings for those exact sources. Bounded to 200 rows per pass per boot (converges over restarts on a larger backlog) and best-effort per row (one failure never aborts the pass). Rows still under the current floor are excluded by a length filter in the SELECT so they are never retried forever, and private entries are excluded from the JOURNALS pass exactly like the live indexing path. * test(rag): scope backfill assertions to their own rows --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
0e9f21de69 |
fix(api): default event loop to asyncio + cancellation-safe commit — kills the CI segfault (#340)
* fix(api): default the event loop to asyncio + cancellation-safe commit The recurring CI e2e segfault traced to uvloop: the harness's uvicorn.run() auto-selected it while production's serve() path never consulted Config.loop (stock asyncio, accidentally safe). Every launch site now resolves ROBOCO_UVICORN_LOOP (default asyncio; uvloop opt-in), and DbCommitMiddleware's commit-in-send can no longer be interrupted mid-wire: on cancellation it gets a bounded grace to finish (committed data survives the 504), else invalidate-and-reraise. * feat(runtime): expected-stop breadcrumbs attribute container deaths Two production exit-143s had no attributable source: every orchestrator kill path now records a short reason breadcrumb, and the exit monitor consumes it -- an expected stop logs its reason at info, a genuinely unexpected one logs none_recorded plus docker-inspect diagnostics (OOMKilled, timestamps) so the next mystery SIGTERM self-identifies. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
312ec990dd |
fix: prod triage 2026-07-08 — MCP auth residue, gateway envelopes, verb-loop cap, A2A interjection, manual spawn UX (#334)
* fix(auth): pass agent UUID to CLI-arg MCP servers (optimal/docs/search) The container token is HMAC-signed over the agent UUID (#314), but the optimal/docs/search MCP servers received the slug as their CLI arg and sent X-Agent-ID=<slug>, so every research/RAG/docs call 401ed with signature mismatch under enforced auth. Pass the already-computed agent_uuid in the three args lists instead. * fix(gateway): include remediate in gateway.rejected audit details Conventions-gate rejections carry the offending file:line listing only in the envelope's remediate field, which the audit row dropped -- ops logs showed just the violation count with no way to see what blocked. * fix(gateway): return envelope on do/commit git failure A GitError from the commit verb propagated to the generic middleware handler, so agents got a raw error blob with no remediate/next. Catch it and return an error envelope; 'no changes added to commit' with an explicit files list now names the mismatch and the omit-files fallback. * fix(agent-sdk): absolute rejection cap breaks slow-drip verb loops The verb circuit breaker only counted rejections inside a 60s sliding window, so an agent retrying i_am_done every 3-4 minutes looped for 30+ minutes without tripping it. Add a session-scoped cumulative per-(verb, task) cap at 3x the windowed limit that trips regardless of pacing. * feat(a2a): CEO chime-in interjects into the viewed conversation Previously reply_as_ceo re-homed the message into a canonical CEO<->target conversation with no panel surface, so a chime-in reported success but was invisible and only opportunistically delivered. interject_as_ceo now inserts the message into the conversation being viewed (from_agent=ceo, directed via an @target content prefix), bumps that conversation's counters with the unread ping keyed to the addressed participant, and both participants see it in transcript and read_a2a. * feat(panel): manual spawn carries task + message, surfaces refusals The agent detail page spawned with no request body (task/message impossible), the spawn button could double-fire (2.5ms double-POST seen live), and refusal reasons never reached the UI: readiness refusals were generic 500s and the already-running no-op looked like success. Detail page now uses SpawnAgentDialog, a synchronous ref guard blocks re-entry, AgentReadinessError maps to 409 with its reason shown, already_running is signalled and toasted, and a task_id builds a task-aware prompt instructing the claim (task_id alone never did), with the CEO's message appended as a note. * test(panel): align a2a page test with the interjection footer copy The chime-in rebuild changed the composer footer; the page-level test asserting the old copy was outside the rebuild's scoped vitest run. * fix(api): commit the request DB session before the response is sent FastAPI unwinds yield-dependencies after the response bytes go out, so get_db's post-yield commit raced the client's next request -- a verb could return ok while its claim/status write was still uncommitted (the e2e ok-without-effect flake family), and a failed commit was silently lost behind an already-sent 200. DbCommitMiddleware (innermost, pure ASGI) commits the session stashed by get_db_committed before forwarding http.response.start; commit failure now surfaces as a 5xx. get_db is untouched for its direct non-request callers. * fix(db): invalidate, not rollback, the session on request cancellation With the commit moved into the send path, the flow-verb timeout can cancel mid-commit; rolling back then issues another command over an asyncpg connection stranded mid-wire-protocol, and the poisoned connection segfaults uvloop/asyncpg when a later checkout recycles it (3/3 identical CI faulthandler dumps). On CancelledError discard the connection via session.invalidate() -- SQLAlchemy's documented handling for a timeout during commit -- and keep rollback for plain exceptions. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
60f571bc02 |
fix(release): publish via GitHub REST — gh CLI is not installed in any image (#331)
ReleaseExecutor.publish_release shelled out to 'gh release create', but no
Dockerfile installs the gh CLI (verified missing in the live orchestrator
container), so an armed release manager died at publish AFTER the release
commit was pushed. Publish now POSTs /repos/{owner}/{repo}/releases with the
project's decrypted token — same auth/httpx pattern as PR creation, same
fail-closed semantics (non-201 -> structured publish_failed, CEO retries;
the 300s deadline is the httpx client timeout). Subprocess publish-timeout
test replaced with REST-path tests (201/non-201/transport-error/no-token).
Co-authored-by: Renn F <rennf93@users.noreply.github.com>
|
||
|
|
0bf0cd69b3 |
fix(release): close the 0.19.0 scan findings — sandbox mongo tag, flow-verb timeout walls, video hardening (#329)
- mongo:8-alpine → mongo:8 (tag never existed; a mongo-opted project could spawn no agents) + a Docker Hub tag-existence e2e guard for every sandbox engine - flow-verb timeouts at both walls: shared SLOW_VERBS policy (i_am_done / submit_up / submit_root / open_pr / i_will_work_on get the 900s server budget); the MCP client now outlasts the server budget (+10s headroom, orchestrator-injected env) so agents receive the middleware's clean 504 envelope instead of dying at the old flat 30s client timeout - cancellation safety: the quality gate kills+reaps its child on CancelledError; create_pr records the PR via a shield-with-wait-out helper so the write can neither be skipped nor race get_db's rollback - video engine: renderer sidecar isolated on a render-only network, 2g/2cpu caps, 570s render watchdog with exit-on-hang, 512MB tar decompression cap, CEO notification on terminal render failure, reject under the approve mutex (fail-closed on Redis-down) - dead python-jose dependency removed (drops ecdsa and its unfixable Minerva advisory PYSEC-2026-1325); panel --font-mono now a real monospace stack Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
2a9d9e25d9 |
feat(tasks): task-content guardrails — structured plans + constraints split (#328)
* feat(tasks): task-content guardrails — structured plans + constraints split Bound task PLANNING content the way journals/notes already are, fixing the poor task quality flagged 2026-07-07 (degenerate roots, over-decomposed leaves, descriptions bloated by an auto-attached conventions dump). Phase A — plan/AC guardrails (no migration): - _pm_sub_tasks_gate: cap sub_tasks at 7; per-subtask ceilings (title <=200, description <=600) enforced at both the Pydantic boundary and the gate. Dropped the min-2-roots and no-subtasks-on-code rules: both contradict the 2026-05-08 rule (test_cell_pm_can_plan_code_typed_parent_via_i_will_plan) and break legitimate single-cell roots. Long comment in the gate explains. - IWillPlanRequest: plan <=2000, approach <=800 (floor 150 kept), typed SubTaskCreate/RiskCreate/OpenQuestionCreate replacing loose list[dict]. - DelegateRequest + task_completeness: acceptance_criteria capped at 7 items, each <=200 chars. New FieldRule.MAX_LENGTH_LIST + _post_rule_reject helper (extracted to keep the gate under xenon B). - Routes dump typed models to dicts for the existing rich_plan shaper. Phase B — conventions split (migration 068): - New nullable tasks.constraints Text column; _attach_baseline_constraints now writes the ## Constraints block there instead of appending to description, so description is the human-authored instruction only. The conventions still reach the agent independently at spawn via the ambient block, so agent correctness is unaffected. - TaskResponse / Task model / panel Task type carry constraints; panel shows a read-only Constraints card. Field is optional on the TS type (backend returns null for flag-off / pre-migration rows). Tests: 5 new gate unit tests, 7 schema tests, 3 AC policy tests, 3 e2e smoke scenarios; 4 baseline-constraints integration tests updated. ruff/mypy/xenon clean; 10026 unit+foundation+e2e green; panel typecheck clean. Refs: plan breezy-imagining-kahn * test(tasks): use typed SubTaskCreate instead of dict literals in plan tests make quality runs mypy over tests/ (1079 files), not just roboco/ — the four sites passing dict literals to the now-typed sub_tasks: list[SubTaskCreate] field failed mypy. Construct SubTaskCreate directly; the typed model raising ValidationError IS the boundary the rejection tests assert. * fix(deps): drop unused python-jose — clears PYSEC-2026-1325 (ecdsa, no fix) CI's pip-audit went red on a freshly-published advisory PYSEC-2026-1325 against ecdsa 0.19.2 (no fix published — 0.19.2 is the latest). ecdsa is a transitive dep of python-jose, which is a DIRECT dep of roboco but is NOT imported anywhere in roboco/ or tests/ (grep-verified). The actual JWT path uses PyJWT (import jwt) + fastapi_users.jwt, not python-jose. So python-jose is a dead dependency. Removing it (deletion over an --ignore-vuln waiver) drops ecdsa + rsa + pyasn1 + their type stubs from the lockfile, eliminating the CVE at the source. deptry roboco/ stays clean (no missing-dep), mypy clean, auth + schema tests pass. Master CI was green 9h before this PR's run, so the advisory published in that window would red any run including master — this fix unblocks both. * chore(prompts): regenerate verb tables for typed plan sub_tasks Phase A's IWillPlanRequest schema change (sub_tasks/risks/open_questions from loose list[dict] to typed SubTaskCreate/RiskCreate/OpenQuestionCreate) made the auto-generated verb tables stale. Regenerated via scripts/regenerate_verb_tables.py — the diff is purely the signature reflection (list[str|str] -> list[SubTaskCreate], etc.). Required by the foundation-check gate (Makefile:559). --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
92ab13bce0 |
Fix/backend/flow verb timeout row lock (#326)
* feat(sandbox): pluggable per-engine registry (postgres/redis/mongo) Replaces the hardcoded postgres+redis branches in the provisioner and the env emitter with a registry of SandboxEngine specs (image, run args, readiness probe, connection, ROBOCO_TEST_* env) in a pure low module (roboco/models/sandbox.py). VALID_SANDBOX_SERVICES is derived from the registry — single source of truth — and the provisioner + orchestrator iterate it, so adding an engine is one class + one registry line, not another branch. Adds a mongo:8-alpine engine (ROBOCO_TEST_MONGO_*) as the third service alongside postgres/redis. Also fixes the cold-pull loop that stranded v0.19.0 board agents with empty error strings: docker run pulled inline under a 20s deadline, so a NAS cold pull was killed, cancelled, and re-pulled from scratch forever. _ensure_image now inspects + pulls (300s) before run; provisioning errors log type+message so a bare TimeoutError no longer shows as "". Panel edit-project dialog: postgres/redis toggles -> a Set<string> multi-select driven by a SANDBOX_SERVICES catalog, so new engines appear in the UI by adding to the catalog. Tests: engine parity (allowlist==registry, unique slugs/images, no None leak in env, SandboxInfo aggregates every engine), mongo provision + env injection, plus the existing postgres/redis provision/env/spawn/janitor suite updated to the registry shape. 821 unit / 5 skip green; ruff + mypy (360 files) clean. * docs(sandbox): reflect pluggable engine registry + mongo across docs CHANGELOG (0.19.0): Added entry for the pluggable sandbox engine registry (postgres/redis/mongo) + Fixed entry for the cold-pull loop/empty-error strand that boarded v0.19.0 board agents. docs/map (9 files): sandbox subsystem blurbs, SandboxProvisioner rows, _maybe_provision_sandbox/_append_sandbox_env rows, feature-flag rows, the migration-057 row + v0.17.0 delta, and the models.md VALID_SANDBOX_SERVICES note — all retitled to DB/Redis/Mongo via the engine registry (roboco/models/sandbox.py), with the one-class-one-line extension story and the _ensure_image cold-pull fix. Production-network (roboco_data) lines left as postgres+redis — mongo is sandbox-only, not a prod service. docs/rag (3 files): sandbox-db.md rewritten around the registry (engine list, generic _provision_engine, image pre-pull, ROBOCO_TEST_DB_*/REDIS_*/MONGO_* incl. MONGO_AUTH_DB=admin, single emit_env); config-reference sandbox flag row + subsection retitled; db-network-isolation framing broadened to postgres/redis/mongo. preconditions-and-rejections left untouched (its hit was an unrelated gateway see-also link). * test(e2e): harden umbrella close terminal reads with bounded wait-for-state The MegaTask umbrella close test flaked once on CI (ceo-approve returned 200 but the re-fetch saw awaiting_pm_review) then passed on re-run. The production path is deterministic: complete -> main_pm_complete -> submit_pm_review -> escalate_to_ceo -> ceo_approve -> commit, all on one session, all awaited; the fire-and-forget completion hooks are isolated (own session, best-effort, never touch task.status or the request session). 20 local runs could not reproduce it. The one real surface is the read pattern: the e2e stack commits on the uvicorn thread's loop and reads via a separate loop (run_db -> asyncio.run with a fresh engine), so a terminal single point-read can race a still-draining completion hook on a contended runner. Replace the two terminal point-reads with a bounded wait_for_status poll. Strictly better than a one-shot read: absorbs the transient, and a genuine state bug still surfaces via the timeout branch asserting against the last-read state. * fix(gateway): bound hung flow-verbs with a server-side timeout A gateway intent-verb whose request transaction held the SELECT ... FOR UPDATE lock on the task row never committed: uvicorn does not cancel the endpoint coroutine on client disconnect and get_db only rolled back on Exception (not a hang/cancellation), so the row lock was held indefinitely and every later task-row write on that task wedged (2026-07-07 kimi-k2.7-code:cloud agent on task 79d686f0). Reads (evidence) and journal writes (note) stayed fast — the symptom that pointed at a task-row lock. Fix: pure-ASGI FlowVerbTimeoutMiddleware wraps each /api/v1/flow/* request in asyncio.timeout(flow_verb_timeout_seconds, default 120s). On expiry the inner app is cancelled; CancelledError now propagates through get_db (which catches it alongside Exception and rolls back), releasing the FOR UPDATE lock, and a retryable 504 gateway_timeout envelope is returned. Pure ASGI (not BaseHTTPMiddleware) so cancellation reaches the route coroutine + get_db dependency directly, with no spawned-task gap. Registered innermost so correlation + logging still wrap the 504. E2E: two fault-injection scenarios in tests/e2e_smoke/test_flow_verb_timeout.py. A hang is injected inside the verb's own transaction (claim acquires the FOR UPDATE lock, then set_plan sleeps past the timeout; only the first set_plan call runs — a retry short-circuits as idempotent re-entry). - ARMED (server timeout 1s): verb-1 returns a bounded 504 gateway_timeout, verb-2 re-acquires the row and reaches the post-claim gate (tracing_gap) — proving the lock was released by verb-1's cancellation. - DISARMED (server timeout 1000s, MCP client timeout 3s): verb-1 holds the lock past the client's HTTP timeout — the empirical reproduction of the wedge on the same branch, by turning the fix off. Full e2e suite green (32 passed). * feat(sandbox): pluggable per-engine registry (postgres/redis/mongo) (#324) (#325) * feat(sandbox): pluggable per-engine registry (postgres/redis/mongo) Replaces the hardcoded postgres+redis branches in the provisioner and the env emitter with a registry of SandboxEngine specs (image, run args, readiness probe, connection, ROBOCO_TEST_* env) in a pure low module (roboco/models/sandbox.py). VALID_SANDBOX_SERVICES is derived from the registry — single source of truth — and the provisioner + orchestrator iterate it, so adding an engine is one class + one registry line, not another branch. Adds a mongo:8-alpine engine (ROBOCO_TEST_MONGO_*) as the third service alongside postgres/redis. Also fixes the cold-pull loop that stranded v0.19.0 board agents with empty error strings: docker run pulled inline under a 20s deadline, so a NAS cold pull was killed, cancelled, and re-pulled from scratch forever. _ensure_image now inspects + pulls (300s) before run; provisioning errors log type+message so a bare TimeoutError no longer shows as "". Panel edit-project dialog: postgres/redis toggles -> a Set<string> multi-select driven by a SANDBOX_SERVICES catalog, so new engines appear in the UI by adding to the catalog. Tests: engine parity (allowlist==registry, unique slugs/images, no None leak in env, SandboxInfo aggregates every engine), mongo provision + env injection, plus the existing postgres/redis provision/env/spawn/janitor suite updated to the registry shape. 821 unit / 5 skip green; ruff + mypy (360 files) clean. * docs(sandbox): reflect pluggable engine registry + mongo across docs CHANGELOG (0.19.0): Added entry for the pluggable sandbox engine registry (postgres/redis/mongo) + Fixed entry for the cold-pull loop/empty-error strand that boarded v0.19.0 board agents. docs/map (9 files): sandbox subsystem blurbs, SandboxProvisioner rows, _maybe_provision_sandbox/_append_sandbox_env rows, feature-flag rows, the migration-057 row + v0.17.0 delta, and the models.md VALID_SANDBOX_SERVICES note — all retitled to DB/Redis/Mongo via the engine registry (roboco/models/sandbox.py), with the one-class-one-line extension story and the _ensure_image cold-pull fix. Production-network (roboco_data) lines left as postgres+redis — mongo is sandbox-only, not a prod service. docs/rag (3 files): sandbox-db.md rewritten around the registry (engine list, generic _provision_engine, image pre-pull, ROBOCO_TEST_DB_*/REDIS_*/MONGO_* incl. MONGO_AUTH_DB=admin, single emit_env); config-reference sandbox flag row + subsection retitled; db-network-isolation framing broadened to postgres/redis/mongo. preconditions-and-rejections left untouched (its hit was an unrelated gateway see-also link). * test(e2e): harden umbrella close terminal reads with bounded wait-for-state The MegaTask umbrella close test flaked once on CI (ceo-approve returned 200 but the re-fetch saw awaiting_pm_review) then passed on re-run. The production path is deterministic: complete -> main_pm_complete -> submit_pm_review -> escalate_to_ceo -> ceo_approve -> commit, all on one session, all awaited; the fire-and-forget completion hooks are isolated (own session, best-effort, never touch task.status or the request session). 20 local runs could not reproduce it. The one real surface is the read pattern: the e2e stack commits on the uvicorn thread's loop and reads via a separate loop (run_db -> asyncio.run with a fresh engine), so a terminal single point-read can race a still-draining completion hook on a contended runner. Replace the two terminal point-reads with a bounded wait_for_status poll. Strictly better than a one-shot read: absorbs the transient, and a genuine state bug still surfaces via the timeout branch asserting against the last-read state. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
8f6dde9a50 |
feat(sandbox): pluggable per-engine registry (postgres/redis/mongo) (#324)
* feat(sandbox): pluggable per-engine registry (postgres/redis/mongo) Replaces the hardcoded postgres+redis branches in the provisioner and the env emitter with a registry of SandboxEngine specs (image, run args, readiness probe, connection, ROBOCO_TEST_* env) in a pure low module (roboco/models/sandbox.py). VALID_SANDBOX_SERVICES is derived from the registry — single source of truth — and the provisioner + orchestrator iterate it, so adding an engine is one class + one registry line, not another branch. Adds a mongo:8-alpine engine (ROBOCO_TEST_MONGO_*) as the third service alongside postgres/redis. Also fixes the cold-pull loop that stranded v0.19.0 board agents with empty error strings: docker run pulled inline under a 20s deadline, so a NAS cold pull was killed, cancelled, and re-pulled from scratch forever. _ensure_image now inspects + pulls (300s) before run; provisioning errors log type+message so a bare TimeoutError no longer shows as "". Panel edit-project dialog: postgres/redis toggles -> a Set<string> multi-select driven by a SANDBOX_SERVICES catalog, so new engines appear in the UI by adding to the catalog. Tests: engine parity (allowlist==registry, unique slugs/images, no None leak in env, SandboxInfo aggregates every engine), mongo provision + env injection, plus the existing postgres/redis provision/env/spawn/janitor suite updated to the registry shape. 821 unit / 5 skip green; ruff + mypy (360 files) clean. * docs(sandbox): reflect pluggable engine registry + mongo across docs CHANGELOG (0.19.0): Added entry for the pluggable sandbox engine registry (postgres/redis/mongo) + Fixed entry for the cold-pull loop/empty-error strand that boarded v0.19.0 board agents. docs/map (9 files): sandbox subsystem blurbs, SandboxProvisioner rows, _maybe_provision_sandbox/_append_sandbox_env rows, feature-flag rows, the migration-057 row + v0.17.0 delta, and the models.md VALID_SANDBOX_SERVICES note — all retitled to DB/Redis/Mongo via the engine registry (roboco/models/sandbox.py), with the one-class-one-line extension story and the _ensure_image cold-pull fix. Production-network (roboco_data) lines left as postgres+redis — mongo is sandbox-only, not a prod service. docs/rag (3 files): sandbox-db.md rewritten around the registry (engine list, generic _provision_engine, image pre-pull, ROBOCO_TEST_DB_*/REDIS_*/MONGO_* incl. MONGO_AUTH_DB=admin, single emit_env); config-reference sandbox flag row + subsection retitled; db-network-isolation framing broadened to postgres/redis/mongo. preconditions-and-rejections left untouched (its hit was an unrelated gateway see-also link). * test(e2e): harden umbrella close terminal reads with bounded wait-for-state The MegaTask umbrella close test flaked once on CI (ceo-approve returned 200 but the re-fetch saw awaiting_pm_review) then passed on re-run. The production path is deterministic: complete -> main_pm_complete -> submit_pm_review -> escalate_to_ceo -> ceo_approve -> commit, all on one session, all awaited; the fire-and-forget completion hooks are isolated (own session, best-effort, never touch task.status or the request session). 20 local runs could not reproduce it. The one real surface is the read pattern: the e2e stack commits on the uvicorn thread's loop and reads via a separate loop (run_db -> asyncio.run with a fresh engine), so a terminal single point-read can race a still-draining completion hook on a contended runner. Replace the two terminal point-reads with a bounded wait_for_status poll. Strictly better than a one-shot read: absorbs the transient, and a genuine state bug still surfaces via the timeout branch asserting against the last-read state. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |