mirror of
https://github.com/rennf93/roboco.git
synced 2026-08-03 07:23:24 +02:00
a036c9798546262536b858cccdcb9e857e333a64
1043
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
|
a036c97985 |
fix(tg): cockpit data correctness — real GLM pricing, display timezone, agent activity tracking (#666)
* fix(tg): cockpit data correctness — real GLM pricing, display timezone, agent activity tracking
Three root causes behind the Mini App/bot showing wrong numbers:
Pricing: glm-5.2 gets a grounded per-token rate (z.ai published pricing,
$1.40/$4.40/$0.26 per 1M, source+date in the table comment) so a GLM
fleet day stops reporting $0.00 for half a million tokens; ungrounded
Ollama-Cloud models render "subscription (untracked)" instead of a bare
zero (is_ollama_cloud_model, consumed directly by the cockpit). Side
effect, intended and documented: honestly-priced GLM now trips the
downgrade-only comparator for new qa/documenter complexity pins.
Display timezone: the cockpit bucketed days in UTC for a GMT+2 operator.
New pure foundation module display_time (resolve_zone/local_date/
trailing_dates/day_bounds_utc, DST-correct with tests for the 23h/25h
days) + ROBOCO_DISPLAY_TIMEZONE (IANA-validated, default UTC); the
cockpit's spend/velocity series bucket raw session/completion rows by
the display zone. The UTC-keyed rollup table and the main dashboard are
deliberately untouched.
Agent activity: AgentTable.status was never set to ACTIVE and
current_task_id was never written anywhere — "active: 0, working: []"
was structurally permanent. Every claim path now marks the claimant
ACTIVE with rollback symmetry (_finalize_claim for dev/PM claims,
_qa_or_doc_claim for QA/doc/PR-gate claims, pr_review_claim for external
review) and every release path clears it (pass/fail QA, pr_pass/pr_fail,
complete_review, advance-to-PM-review, reaper unclaim, voluntary
unclaim, reassign retarget, pool divert, admin transitions, unblock
restore-to-in-progress). The bot's /status shares the cockpit's fleet
derivation so the two surfaces can't disagree. Known ceiling, commented:
one current_task_id column shows a multi-root coordinator PM's most
recent claim only.
Drill: sonnet develop -> sonnet adversarial (refuted the original
chokepoint coverage claim; QA/doc/reviewer paths were unwired) ->
correction round (wired them all + restored a dropped assertion, deleted
a dead helper and the dead subscription_billed field) -> review.
* fix(db): post_update on AgentTable.current_task breaks the flush cycle
agents.current_task_id and tasks.assigned_to reference each other, so a
flush touching both rows — every claim now marks its agent ACTIVE — is
an instance-level circular dependency SQLAlchemy cannot topologically
sort. The e2e smoke's full verb paths (12 tests) hit it; the unit and
integration suites never flush both dirty rows with relationships
loaded. post_update emits the FK as a second UPDATE, the canonical fix
for mutually-referencing rows.
* fix(budgets): enforce only explicitly-set budgets — no per-TaskType defaults
The per-TaskType default cap table blocked an unbudgeted coordination
root one opus planning turn in ($1.50 PLANNING default vs. real
coordination spend) — a false positive by design the moment the fleet
runs a priced model. Budgets are now explicit-input only:
effective_task_budget_usd returns None for an unset budget_usd, the
budget sweep skips enforcement (and never prices spend) on None, and
the unblock re-check passes on None so clearing the budget field is
itself a valid resolution. The project monthly cap stays as the
explicit-input fleet-wide backstop. Panel copy tells the truth
("No cap" placeholder; empty = uncapped), and the TaskType default
table plus its resolver are deleted.
---------
Co-authored-by: Renn F <rennf93@users.noreply.github.com>
|
||
|
|
0c1d450a05 |
fix(intake): emit the complete reply when no text deltas streamed (#665)
The intake/secretary chat driver treated StreamEvent text deltas as the ONLY text channel: AssistantMessage TextBlocks were always swallowed as already-streamed (the double-render guard). The CLI's partial-message emission turned out to be remotely gated — on 2026-07-23 the NAS containers got zero stream_event lines from the identical binary, flags, SDK, model, and settings that stream fine elsewhere — so the guard became a total blackout: replies were generated, the relay carried only init/status/turn_end, and the CEO saw nothing. SdkIntakeSession.send now tracks whether any text delta arrived during the turn; normalize() emits an AssistantMessage's complete text only when none did. Streaming mode is byte-identical (deltas render live, completes stay suppressed); gated mode delivers the reply as one block instead of nothing. Covers Secretary (same machinery). Regression tests: fallback emission, default suppression, and both modes end-to-end through the session layer. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
d87ce2ca11 |
docs(changelog): document #663 (mix-mode escape hatch + per-loop DB engine) (#664)
The 0.27.0 release proposal executes its STORED drafted changelog, and the current proposal predates #663 — approving it as-is would ship a release whose commit range includes both #663 fixes with no changelog line, which the next readiness sweep would then flag as curation gaps. Documenting them in [Unreleased] so a re-originated proposal drafts complete; verified against the gap-check matching rules (every required commit in v0.26.0..HEAD matches by #PR or exact summary). Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
Renzo FandGitHub
|
226e1b586a | fix(routing): escape hatch out of mix mode — clear-all overrides + pins warning (#663) | ||
|
Renzo FandGitHub
|
dd4c3c3ed0 | docs(release): prepare 0.27.0 — curated changelog, rag/map sweep, concat rebuild script (#662) | ||
|
Renn F
|
d8841279a8 | Updated uv.lock | ||
|
Renzo FandGitHub
|
d4b7e1e7b8 | fix: post-finale completeness sweep — routing surface, provider config, budgets, compose env, interactive exemption (#661) | ||
|
|
21d6730400 |
feat(providers): Gemini CLI provider — ModelProvider.GEMINI (#660)
* feat(providers): Gemini CLI provider — ModelProvider.GEMINI Mirrors the grok blueprint with source-verified divergences (all facts pinned against google-gemini/gemini-cli @ 9681621c): no refresher daemon — Google's refresh tokens are reusable, so the RO host mount is COPIED to a writable container-local ~/.gemini and each container refreshes in-process independently (the write-back crash risk on RO never triggers); settings.json renders security.auth.selectedType 'oauth-personal', experimental.enableAgents=false (subagent ban), autoConfigureMemory=false with a bounded heap; tool scoping rides the tiered TOML Policy Engine (deny-only rules that yolo mode structurally cannot beat); gemini -p with --output-format stream-json; usage parsed from the run's own stdout stats — the adversarial pass caught the parser reading the json-mode nested shape while the entrypoint runs stream-json's FLAT shape (every real run would have priced $0 forever, hidden by fixtures sharing the assumption) — now flat-primary with the nested shape as cited fallback; rate-limit classified from structured error.type only (model-echo immune), native exit 41 auth passthrough; per-model pricing for the three GA models; migrations 084 (enum) + 085 (seed) complete the 082-085 finale chain. V1 excludes interactive intake/secretary. Stack-merge required two behavior-preserving complexity refactors in the shared park/usage plumbing (a park-pair loop; a usage-reader dispatch dict). * fix(providers): route gemini usage read through the containment barrier Mirrors the codex/grok fix — _gemini_usage_json now delegates to _read_usage_json_contained, so CodeQL's path-injection alert on the gemini read is resolved by the same resolve-and-contain guard. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
Renn F
|
13abb2ece0 |
fix(docker): install codex CLI via npm, not the CDN-denied installer
The chatgpt.com/codex/install.sh installer returns 'resource is denied' to non-browser clients in CI, breaking the agent-codex image build. Switch to npm install -g @openai/codex@<pinned>: the package has no postinstall network fetch (verified) — its native binary rides an optionalDependency served from the public npm registry, and node is already in the base image. Verified locally: install succeeds, codex --version runs. |
||
|
|
c70ff3cf9a |
feat(providers): Codex CLI provider — OpenAI via ModelProvider.OPENAI (#659)
* feat(providers): Codex CLI provider — OpenAI via ModelProvider.OPENAI Mirrors the grok blueprint end to end: CodexCliProvider (RO ~/.codex mount, ANTHROPIC_* blanked), an orchestrator-side codex_auth.py refresher (JWT-exp staleness, atomic rewrite, lock-serialized single-use rotation, --check backstop; the CLI's own in-process refresh write no-ops on the RO mount by design — margins keep the orchestrator ahead of the CLI's 5-minute window), config.toml rendering with required=true gateway MCP servers, execpolicy deny rules (forbidden-only), per-role --sandbox (developer=workspace-write, review/doc roles read-only), codex exec --json with pinned ROBOCO_CODEX_CLI_MODEL (gpt-5.3-codex), usage summed from typed turn.completed events priced via the real 4-bucket split, dedicated image + entrypoint, registry/park/finalize/ compose/release wiring. V1 excludes interactive intake/secretary. Per adversarial review: migration 083 seeds the openai provider row enabled=True (without it every routing path 404'd — the whole feature was operationally dead code; grok needed the same seed in 039), the panel picker gained the OpenAI catalog group it silently lacked, and exit classification is structural — only stderr and error.message fields from error events are sniffed (word-boundaried patterns, exact auth phrases, bare 'login' dropped), so the model echoing on-topic words can never false-park the provider fleet-wide, proven by a benign-transcript test. Known open risk flagged, not claimed: whether codex's workspace-write OS sandbox excludes /app is unverified, and no hook mechanism exists to port the bash-guard defense-in-depth. * fix(providers): containment barrier on usage.json reads (code scanning) CodeQL flagged the codex usage read as path injection — correctly: os.path.basename does not neutralize '..', and the upstream segment validator isn't in CodeQL's taint model. The grok/codex reads collapse into one _read_usage_json_contained helper that resolves the built path and refuses anything outside the resolved usage root — a hostile id can never escape regardless of upstream drift. Traversal + containment regression tests added; a stray noqa in the test file replaced with a named constant per repo rule. * fix(providers): use realpath+startswith containment CodeQL recognizes The is_relative_to() guard was a real barrier but not in CodeQL's py/path-injection sanitizer model, so the alert persisted. Switch to the canonical os.path.realpath + startswith(root + os.sep) form, which CodeQL recognizes as a path-traversal barrier; behavior is identical (refuse any candidate resolving outside the usage root). * fix(providers): regexp-allowlist the usage-id segment (CodeQL barrier) Neither is_relative_to nor realpath+startswith was recognized by CodeQL's py/path-injection sanitizer model across the str->Path->open flow. Sanitize the tainted component at the source instead: the id must fullmatch a strict slug token ([A-Za-z0-9][A-Za-z0-9._-]*, no separators, no '..'), which CodeQL recognizes as a path-injection barrier; the realpath+startswith containment stays as defense-in-depth. * fix(providers): standalone regexp guard so CodeQL recognizes the barrier The sanitizer was one disjunct of a compound 'or' condition, which CodeQL's guard analysis does not trace as a barrier. Split the regexp fullmatch into its own single-condition guard (the redundant '..' check is dropped — the required alphanumeric first char already excludes it). --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
165892dc62 |
feat(routing): cost-tiered complexity routing + saved presets (#656)
The 08-31 lever: model_assignments gains one compound rung —
AGENT_SLUG > ROLE('{role}:{complexity}') > ROLE > GLOBAL — so a
low-complexity task can route to a cheaper tier while coordinators stay
pinned. Structurally opt-in: zero rows means byte-identical routing
(pinned by a named test across every precedence case), the cost_tiered
apply-mode (seeds developer:low→haiku) is reachable only from the
explicit PM-gated endpoint — verified no startup path can apply it.
Overrides are downgrade-only (input-price comparator), allowlisted to
{developer, qa, documenter} — cell_pm excluded per the org's own
coordinator definition and its documented weak-model incidents — and
validated at write time (disabled/unconfigured provider rejected with
remediation; cross-provider-family overrides warn explicitly).
Per adversarial review: the four mode-switch applies now spare compound
rows exactly like agent pins (the 2026-07-17 unscoped-wipe class, new
victim, same fix extended via one shared wipe helper) with panel cache
invalidation + truthful confirm dialogs; preset apply validates the
entire payload BEFORE the wipe (validate-all-first), with a savepoint
crash test proving rollback.
Presets (CEO request): routing_presets table (migration 082) snapshots
the full mix — mode, per-agent overrides, complexity rows — with
save/apply/delete endpoints and a panel preset bar; applying skips
since-removed models with per-entry notes, never silently.
Task complexity threads task_id through _resolve_agent_route at both
call sites; taskless spawns unchanged. 235 backend + 23 panel tests.
Co-authored-by: Renn F <rennf93@users.noreply.github.com>
|
||
|
|
10f039c36f |
feat(eval): golden-task eval harness + doctrine cohort stamp (#655)
* fix(notifications): exponential backoff + CAS claim for expired-unacked re-escalation The sweep re-escalated every expired unacked ack-required notification on every ~60s tick, forever — the live incident: 3 fresh blocker escalations + Telegram DMs per minute from a static stale pile. Now each notification carries reescalation_count / last_reescalated_at / reescalation_delivered_count (migration 079): first fire at expiry, then doubling intervals from 1h capped at 24h, hard stop after ROBOCO_NOTIFICATION_MAX_REESCALATIONS (default 5) with one permanent log carrying attempts-vs-delivered so 'seen and ignored' is distinguishable from 'route never worked'. The due/wait/capped decision is a pure function in foundation/policy/communications.py. Per adversarial review, the attempt slot is claimed by compare-and-set (UPDATE ... WHERE reescalation_count = :n) BEFORE delivery — the previous draft leaned on the 60s dedup window, which never engages for BLOCKER_ESCALATION (_LOOP_PRONE_TYPES excludes it), so concurrent sweeps would have double-delivered. A lost claim skips delivery outright. Legacy rows read as count=0 and keep today's first-fire semantics. 61 tests incl. a two-session CAS race and a real alembic upgrade/downgrade round trip. * feat(budgets): per-task and per-project cost budgets (flag-gated) tasks.budget_usd + projects.monthly_budget_usd (migration 080, chained on 079; adds ix_agent_spawn_sessions_task_id since both enforcement seams filter on bare task_id). Behind ROBOCO_TASK_BUDGETS_ENABLED (default off, feature-flags card) — verifiably inert when off. Claim-time: a project-month-spend guard applies to WORK-STARTING claims only (i_will_work_on / i_will_plan) — per adversarial review, review/ doc/gate/inbound-PR claims are exempt so in-flight work can always finish reviewing and merging at cap. Spend counts closed sessions' estimated_cost_usd PLUS open sessions priced live from token snapshots (the original closed-only sum read parallel long sessions as $0). Sweep-side: the existing budget sweep also prices the active task's spend vs budget_usd (TaskType defaults when null); on breach the task is BLOCKED (HUMAN resolver, budget marker) BEFORE the graceful stop so the unclaim no-ops and the dispatcher never respawns onto it, and the CEO notification names both recovery steps. unblock on a budget-blocked task re-checks live spend and refuses while still over — no silent re-breach loop. Panel: budget inputs in both dialogs (0 rejected — a zero budget silently blocks everything), spend logic consolidated in TaskService.task_spend_usd. 42 new tests incl. a real-DB spend-query suite and a two-tick non-refire sweep test. * feat(eval): golden-task eval harness + doctrine cohort stamp roboco/eval: 6 BenchTaskSpec fixtures run through the real lifecycle in a disposable environment (the e2e_smoke harness's fake GitHub + local git origin + throwaway DB catalog — real isolation, not convention), scored deterministically (terminal status, revision_count, cycle time, tokens/cost via the agent_spawn_sessions task_id join) plus a local- model judge whose output is nested under a non_deterministic-marked object so cohort diffs don't read judge noise as regression. CLI: python -m roboco.eval run --role <slug> --cohort <name>. Source- checkout-only by declared posture (deptry-scoped ignore + a hard ImportError guard naming why; tests/ never ships in images or wheels). agent_spawn_sessions.doctrine_version (migration 081, chained on 080) is stamped at spawn-session finalize from the composed prompt layers — with the session's model column it identifies a cohort durably. Per adversarial review: bench runs patch the vault flags off (they were writing real markdown into the operator's vault), and the real-spawn OrchestratorStageSpawner is deliberately cut to NotImplementedError — spawned containers' MCP wiring resolves to the production orchestrator under real agent UUIDs, so real spawns wait for a dedicated follow-up; the injectable scripted spawner is the working path. Full suite 13852 passed / 94% coverage in the source worktree; deptry/mypy/xenon clean. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
7c8453e210 |
feat(budgets): per-task and per-project cost budgets (flag-gated) (#654)
* fix(notifications): exponential backoff + CAS claim for expired-unacked re-escalation The sweep re-escalated every expired unacked ack-required notification on every ~60s tick, forever — the live incident: 3 fresh blocker escalations + Telegram DMs per minute from a static stale pile. Now each notification carries reescalation_count / last_reescalated_at / reescalation_delivered_count (migration 079): first fire at expiry, then doubling intervals from 1h capped at 24h, hard stop after ROBOCO_NOTIFICATION_MAX_REESCALATIONS (default 5) with one permanent log carrying attempts-vs-delivered so 'seen and ignored' is distinguishable from 'route never worked'. The due/wait/capped decision is a pure function in foundation/policy/communications.py. Per adversarial review, the attempt slot is claimed by compare-and-set (UPDATE ... WHERE reescalation_count = :n) BEFORE delivery — the previous draft leaned on the 60s dedup window, which never engages for BLOCKER_ESCALATION (_LOOP_PRONE_TYPES excludes it), so concurrent sweeps would have double-delivered. A lost claim skips delivery outright. Legacy rows read as count=0 and keep today's first-fire semantics. 61 tests incl. a two-session CAS race and a real alembic upgrade/downgrade round trip. * feat(budgets): per-task and per-project cost budgets (flag-gated) tasks.budget_usd + projects.monthly_budget_usd (migration 080, chained on 079; adds ix_agent_spawn_sessions_task_id since both enforcement seams filter on bare task_id). Behind ROBOCO_TASK_BUDGETS_ENABLED (default off, feature-flags card) — verifiably inert when off. Claim-time: a project-month-spend guard applies to WORK-STARTING claims only (i_will_work_on / i_will_plan) — per adversarial review, review/ doc/gate/inbound-PR claims are exempt so in-flight work can always finish reviewing and merging at cap. Spend counts closed sessions' estimated_cost_usd PLUS open sessions priced live from token snapshots (the original closed-only sum read parallel long sessions as $0). Sweep-side: the existing budget sweep also prices the active task's spend vs budget_usd (TaskType defaults when null); on breach the task is BLOCKED (HUMAN resolver, budget marker) BEFORE the graceful stop so the unclaim no-ops and the dispatcher never respawns onto it, and the CEO notification names both recovery steps. unblock on a budget-blocked task re-checks live spend and refuses while still over — no silent re-breach loop. Panel: budget inputs in both dialogs (0 rejected — a zero budget silently blocks everything), spend logic consolidated in TaskService.task_spend_usd. 42 new tests incl. a real-DB spend-query suite and a two-tick non-refire sweep test. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
1d5a8e846f |
fix(notifications): exponential backoff + CAS claim for expired-unacked re-escalation (#652)
The sweep re-escalated every expired unacked ack-required notification on every ~60s tick, forever — the live incident: 3 fresh blocker escalations + Telegram DMs per minute from a static stale pile. Now each notification carries reescalation_count / last_reescalated_at / reescalation_delivered_count (migration 079): first fire at expiry, then doubling intervals from 1h capped at 24h, hard stop after ROBOCO_NOTIFICATION_MAX_REESCALATIONS (default 5) with one permanent log carrying attempts-vs-delivered so 'seen and ignored' is distinguishable from 'route never worked'. The due/wait/capped decision is a pure function in foundation/policy/communications.py. Per adversarial review, the attempt slot is claimed by compare-and-set (UPDATE ... WHERE reescalation_count = :n) BEFORE delivery — the previous draft leaned on the 60s dedup window, which never engages for BLOCKER_ESCALATION (_LOOP_PRONE_TYPES excludes it), so concurrent sweeps would have double-delivered. A lost claim skips delivery outright. Legacy rows read as count=0 and keep today's first-fire semantics. 61 tests incl. a two-session CAS race and a real alembic upgrade/downgrade round trip. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
31a489b431 |
chore(packaging): make quickstart — one-command registry bring-up + pull-smoke CI (#653)
scripts/bootstrap.sh: idempotent bring-up for the pull-and-run deploy. Fresh .env: copies .env.example and injects the three required secrets using the documented one-liners (the panel token via the exact HMAC formula issue_panel_token uses), with a standing-credential warning — and a louder one when cloud auth is detected — since compose's :? guard refuses an empty token unconditionally (verified live). Reused .env: never touched, but the three required vars are pre-validated with pointed remedies instead of compose's opaque interpolation error. Then pull + up -d + a doctor-style readiness sweep grounded in the real surfaces (root /health, /api/auth/status through nginx, the verbatim 'Alembic upgrade finished' log line, ollama list), each stage failing loud with the command to run next. Exposed as make quickstart; README leads with it and keeps the manual steps as 'what quickstart does'. Also found and fixed along the way: the documented registry quickstart was already broken — nginx's :?-required ROBOCO_PANEL_AGENT_TOKEN ships empty in .env.example, so the 4-step path failed at compose config. release.yml gains a pull-smoke job (fresh runner, own GHCR login, needs publish-images): literally pulls the registry compose against the just-published tag, guarding the missing-image regression class that already happened once. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
fa459998b4 |
feat(git): env-ladder rung protection at the shared remote-delete chokepoint (#651)
Rung protection lived only in delete_task_branch; the post-merge PR- source cleanup (and the stale-branch sweep's shared primitive) could still delete a branch that IS a ladder rung. _protected_branches_for_ deletion(slug) — field ∪ rung names, null-ladder shim included — now feeds _delete_remote_branch_best_effort, so every remote deletion path is covered; delete_task_branch's local rung check is removed as exactly subsumed (verified byte-identical comparison semantics). Bonus closed gap: a renamed trunk (default_branch 'trunk', null ladder) is now delete-protected, which the hardcoded main/master floor never covered. Per adversarial review, the deletion lookup fails CLOSED: a raised project lookup skips the delete with a warning (a skipped best-effort delete just retries next sweep — free safety), while a genuinely-gone project proceeds with the hardcoded floor (its ladder is meaningless). The rebase/sync resolver stays fail-open — a refused rebase on a DB blip would wrongly block work, a different tradeoff, now documented. Panel tooltip updated to the new truth. 29 tests. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
3806317aa7 |
fix(guard): operator-scoped XFF hop peers + tailnet allowlist (live-incident fix) (#650)
Two coupled hardenings from the chain-peers adversarial rounds plus the
root-cause fix for the live post-deploy incident where the CEO was
blocked from the panel ('IP not allowed: 100.x.x.x').
Hop peel-set: the whole docker bridge pool leaves the XFF hop set — hops
are now loopback plus operator-named single addresses only
(ROBOCO_GUARD_TRUSTED_CHAIN_PEERS, plain IPs; CIDR entries rejected with
a warning because a range readmits sibling containers). Default-empty
closes the CGNAT-forge residual outright; a gateway-fronted Tailscale
Serve deploy sets its real gateway IP, and a rate-limited detection log
names exactly that IP when an unconfigured host-proxied tailnet chain is
seen, so the silent-regression shape is observable. The connecting-peer
gate (may nginx present XFF at all) deliberately keeps the broad bridge
pool — different check, unchanged.
Incident root cause: guard-core's whitelist is an EXCLUSIVE allowlist
(any non-member is refused), so honestly resolving the tailnet client IP
made ip_security reject the CEO. The tailnet CGNAT range joins
_guard_whitelist() deliberately: Tailscale authenticates device
membership before a packet arrives, real-IP stamping still buys correct
attribution, and any future non-tailnet exposure keeps full scrutiny.
Both compose files now pass ROBOCO_GUARD_EMERGENCY_WHITELIST through to
the orchestrator (the operator escape hatch previously did nothing in a
compose deploy).
NAS is running ROBOCO_GUARD_PASSIVE_MODE=true as interim mitigation —
flip back to false when this deploys. 66 tests.
Co-authored-by: Renn F <rennf93@users.noreply.github.com>
|
||
|
|
da4d9b333d |
feat(git): protected-branches enforcement + panel editor (#649)
projects.protected_branches existed end-to-end but nothing consulted it — the panel had no editor and the git safety checks used hardcoded sets. Now: GitService._protected_branches_for(slug) (frozenset, stripped, fail-open to the hardcoded floor with a warning log) is unioned — never replacing, only tightening — into rebase()'s refusal set, the shared _delete_remote_branch_best_effort skip set (threaded through every caller: task cleanup, PR merge/close cleanup), and sync_task_branch, which now refuses to force-push a protected-named head (the dev-facing sync_branch verb path the HTTP-only fix would have missed). Matching is exact and case-sensitive; an empty list degrades to exactly the old hardcoded behavior, pinned by union-floor regression tests (master/main stay refused regardless of the project list). Panel: chips editor for the field in the edit-project dialog (add via Enter/comma, paste-splitting on comma-separated lists, dedup, clear-to- empty persists []) with an honest tooltip scoped to what is actually enforced. Tests cover both the incumbent GitHub-App dialog suite and the new Protected Branches suite in one harness. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
4585a248ce |
feat(x): redraft loop on CEO reject — feedback re-enters the draft flow (#648)
* feat(x): redraft loop on CEO reject — feedback re-enters the draft flow A rejected X draft's reason used to die with the cancel. reject() with a non-blank reason now schedules a redraft after its commit (defer_after_commit; fresh session; never blocks or fails the HTTP response): XEngine.redraft_from_rejection re-drafts the same source kind via the local model with the reason and rejected body folded in as revision guidance, originating ONE fresh held draft — mirroring the video pipeline's reauthor_from_rejection. Local-model failure or empty output originates nothing (no degraded copies); markers carry forward whole so a redrafted reply/spotlight stays fully functional downstream; bodies ride the same 280 clamp; the open-posts cap holds. Hardened per adversarial review: reject() is now idempotent on an already-CANCELLED target at both check sites (mirroring approve's already_rejected guard — a replayed reject schedules nothing), and the dedup check+originate runs under a non-blocking identity-keyed Redis lock (SET NX + compare-and-del, matching the approve/reject mutex style) so racing rejects can't stack duplicate drafts — lock held or Redis down skips the redraft, which is always safe. Tests pin the fresh-session contract by session identity, the replay no-op, the lock-skip, and clean up their own committed rows. * fix(tests): runtime UUID import + typed task-id coercion in x cleanup helper CI's quality gate runs mypy over tests/ (the local pass covered only roboco/): the _delete_tasks calls handed ORM-typed ids where uuid.UUID was expected. Coercing at the call sites then exposed that UUID was imported under TYPE_CHECKING only — a runtime NameError. Import moved to runtime; both call sites coerce explicitly. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
d1f9d21a68 |
fix(tg): mini-app responsiveness — safe-area activation, truncation, touch targets (#647)
Root cause first: no viewport export existed anywhere, so viewport-fit was never 'cover' and every env(safe-area-inset-*) resolved to 0 on notched iPhones — content under the status bar, dock without real home-indicator clearance. The export lives on the (tg) group layout (server component), NOT app-wide: the dashboard shell has no safe-area padding and must not inherit cover. Also: min-w-0 on four truncating flex children that overflowed their justify-between rows (chat names/previews, fleet task titles); object-contain on the approvals video (letterbox instead of distort on short phones); break-words on the changelog pre / task description / quoted mention; touch targets bumped to >=36px (sheet close, segmented controls, ack button, chips, back button, bell, jump-to-latest, cut-toggle); overflow-x-hidden backstop on the (tg) main scroller; fleet avatar strip sliced to 3 with a +N badge instead of silent clipping. Verified: pnpm typecheck clean, lint 0 errors, panel suite 870/870. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
583d43f213 |
fix(guard): resolve the tailnet client behind host-proxy hops for the WAF (#646)
fastapi-guard peels a fixed trusted_proxy_depth=1 from X-Forwarded-For (the rightmost entry, which nginx itself recorded). That is correct for every chain except host-proxied tailnet traffic (Tailscale Serve → nginx), which arrives as [tailnet-client, loopback-or-bridge-gateway] — depth-1 resolves it to a whitelisted hop IP, leaving WAF/ban/rate-limit inert for the whole /tg surface (the documented ceiling). ClientIpResolutionMiddleware (pure ASGI, wraps SecurityMiddleware so it runs first) stamps guard_core's request.state.client_ip cache — its supported pre-resolution seam — for EXACTLY that shape: peel known local hops (loopback + docker bridge pool) from the right, stamp only when at least one hop was peeled AND the candidate is in the tailnet CGNAT range (100.64.0.0/10). Every other shape abstains, so direct LAN clients, agent containers relaying through nginx (even with forged public-IP prefixes), and all-hops operator traffic resolve byte-for-byte as before. Documented residual: a same-bridge container forging a CGNAT prefix only DE-privileges itself (loses its whitelist exemption). XFF is read first-occurrence to match Starlette's own header semantics, and a wiring test pins the middleware mount ORDER, not just presence. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
98a96bcd21 |
chore(backup): env-gated off-disk mirror + restore drill doc (#645)
The pg_dump sidecar wrote its dumps to the same disk it protects — one disk failure lost both. Setting ROBOCO_BACKUP_MIRROR_DIR in .env to a path on a different disk (external/remote mount) arms a mirror step after every successful dump: tmp+rename copy, mirror pruned to the same BACKUP_KEEP, unwritable mirror logs-and-skips without blocking the primary. Unset, the script never attempts a copy — no fake off-disk copies on the same disk. Docs gain the mirror setup and a quarterly restore drill (throwaway pgvector container, pg_restore, row-count sanity check). Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
2e889c7009 |
feat(lifecycle): inherit advanced upstream base on work re-claims (#644)
* feat(lifecycle): inherit advanced upstream base on work re-claims A re-claim reused a branch cut at an earlier claim, so upstream work merged since (UX/UI landing on the root after the cell branch was cut) never reached BE/FE branches — divergence and avoidable conflicts. _finalize_claim now merges the advanced base into the pre-existing branch via the dependency-lineage merge: already-ancestor is a no-op, a conflict aborts at the cut point and leaves a transition note steering the agent to sync_branch, a clean merge logs an audit trail; never fails the claim. Double-gated: by role (developer/cell_pm/main_pm — QA/documenter/gate claims review the branch as pushed and never move it) AND by pre-claim status (pending/needs_revision only — a PM's i_will_plan re-claim of its own awaiting_pm_review task must not move a branch that already passed QA + the PR gate). Fresh cuts already branch from the live remote base. Cell-PM prompt now orders reading the upstream design docs before planning. * fix(lifecycle): extract base-inheritance gate predicate for xenon budget The four-condition inline gate pushed _finalize_claim to cyclomatic rank C; the quality gate caps blocks at B. The decision moves to a pure module-level predicate, byte-for-byte the same logic. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
b91229f487 |
fix(orchestrator): break the notification-driven respawn loop (#643)
The escalation/approval dispatchers spawn a notification's recipient every cooldown window for as long as it stays pending. These spawns carry no task_id, so the PM respawn breaker never sees them — a single wedged alert/escalation whose recipient never resolves it respawns that recipient forever. Observed live: fe-pm's unacked alerts kept main-pm/fe-pm spawning every ~2-3 min for 6+ hours. Two guards, both gating the spawn after the existing cooldown: - A hard per-(agent, notification) attempt cap (notification_spawn_max_attempts, default 5): once a notification has respawned its target that many times without being acknowledged, stop and log once. The count is id-scoped and survives map pruning (re-stamp), so a fresh escalation is unaffected. - A live-work check before spawning: skip when the notification has expired, is stale past notification_spawn_max_age_seconds (default 6h — wedged or reloaded from before a restart), or its related task is already terminal. Fail-open — a failed task fetch or unparseable field never suppresses a real escalation. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
Renn F
|
f74131a122 |
fix(release): commit the approve() COMPLETED write under the release lock
approve() flushed the published proposal's COMPLETED status but left the durable commit to the background caller (_run_approve_background), which runs after approve()'s finally has already released the Redis lock. In that window a concurrent reject() could acquire the freed lock, re-read a row whose COMPLETED write was only flushed (invisible to its own session under READ COMMITTED), pass its guard, and flip the just-published proposal to CANCELLED — last writer winning the row. The bug-sweep (#638) fixed reject()'s side of this but left approve()'s, so reject()'s 'fails closed' guarantee didn't hold end to end. Commit COMPLETED while still holding the lock (mirroring XPostService._post), so it's durable before release and a racing reject sees it and refuses. A cross-session regression test proves a fresh connection sees COMPLETED the moment approve returns (it read 'pending' before the fix). |
||
|
|
17de29545a |
[6788ce7f] Silent bug sweep: concurrency, state integrity, engine edge-cases, panel data freshness (#638)
* [943d8c4d] Frontend data freshness and approval-queue reliability audit (#631)
* [233a8b0f] WebSocket reconnect message-loss audit and fix (#625)
* [233a8b0f] fix(panel): add REST catch-up to useNotificationStream on WS reconnect
connection.ts has no message buffering/replay, so a notification published
while the CEO bell's socket was down (disconnected/reconnecting) was lost
forever instead of merely delayed. Add a reconnect-triggered GET
/notifications?unread_only=true catch-up folded into the existing
notification_id dedup so a notification delivered both via catch-up and
live WS is never double-counted, and make clearMessages drop the held
catch-up batch too. use-a2a-live.ts and use-rate-limit-websocket.ts were
audited and already have working reconnect-triggered REST fallbacks
(verified via a2a/page.tsx, rate-limit-banner.tsx, usage-overview-panel.tsx
and their existing F083 tests) so no fix was needed there.
* [233a8b0f] docs(panel): add comprehensive WebSocket hooks reference and reconnect architecture guide
Add panel/docs/frontend/hooks.md with full API reference for useWebSocket, useNotificationStream (with new REST catch-up behavior), useAgentStream, useA2ALiveStream, and useConnectionStatus. Include examples, best practices, and testing guidance.
Add panel/docs/architecture/websocket-reconnect.md documenting the message-loss mitigation pattern: Strategy 1 (REST catch-up for events, used by useNotificationStream) and Strategy 2 (REST invalidation for state, used by A2A/rate-limit consumers), plus the dedup logic ensuring no notification is double-counted on reconnect.
---------
Co-authored-by: Frontend Developer 1 <fe-dev-1@roboco.tech>
Co-authored-by: Frontend Documenter <fe-doc@roboco.tech>
* [d5315683] fix(frontend): add distinct toast feedback for silently-swallowed x-post and release-proposal statuses, plus regression tests for all 4 approval queues (#626)
Co-authored-by: Frontend Developer 2 <fe-dev-2@roboco.tech>
* [cd953838] Data-hook null-guard audit and API client 429 retry-by-method fix (#630)
* [cd953838] fix(panel): gate 429 retry by HTTP method, add hook null-guard regression tests
* [cd953838] chore(conventions): waive test-fixture wrapper in hooks null-guard test
* [cd953838] docs(frontend): document API rate-limit retry behavior and null-guard audit results
Added `docs/frontend/api-rate-limiting.md` to document the 429 retry strategy: GET/PUT auto-retry, POST/PATCH/DELETE require X-Idempotency-Key header. Updated `docs/frontend/hooks.md` to confirm the data-hook null-guard audit found all hooks already have correct `enabled` guards and include a regression test suite for the board-review poll on/off behavior and enabled-guard assertions.
---------
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 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>
* [4534c71a] Backend concurrency, state-machine, and engine audit (#634)
* [41de844a] fix(lifecycle): sync CLAIM_RULES with runtime + clear stale claimant on PM hand-off (#627)
Two confirmed state-machine gaps found while auditing lifecycle.py,
task_lifecycle.py, the _ESCALATABLE_TO_BLOCKED bypass, and every
_REVIEW_QUEUE_STATES entry point:
- lifecycle.py's CLAIM_RULES/claim-ActionSpec/StatusTransition table
did not grant CELL_PM/MAIN_PM re-claim of AWAITING_PM_REVIEW even
though task.py's runtime _ROLE_CLAIM_STATUSES already granted it
and claimed the spec agreed -- the two tables had silently drifted,
breaking i_will_plan re-claim on an awaiting_pm_review task.
- docs_complete's _maybe_advance_to_pm_review pre-assigns a specific
owning PM via assigned_to but left claimed_by/active_claimant_id
pointing at the outgoing documenter, unlike every sibling transition
into a review-queue state. A stale active_claimant_id makes
content_actions.py's _active_claim_violation wrongly reject the
newly-assigned PM's own content writes before it formally claims.
Reassign claimed_by + active_claimant_id to the owning PM alongside
assigned_to.
Adds a regression test asserting the documenter's stale claim does not
survive the docs_complete -> awaiting_pm_review hand-off.
Co-authored-by: Backend Developer 2 <be-dev-2@roboco.tech>
* [0c46f666] Engine dedup race + sequencing.py edge-case audit (#628)
* [0c46f666] fix(sequencing): dedup race audit + collision-edge fallback bug
Audited the list-open-then-originate dedup pattern across six engines:
RoadmapEngine, XEngine.run_cycle, DepUpdateEngine, and CIWatchEngine each
run inside exactly one sequential orchestrator-loop asyncio task (no other
call site invokes run_cycle), so they cannot race with themselves; their
in-cycle dedup sets/keys are correctly built before any commit. SelfHealEngine
is the same shape. VideoEngine.open_video_task is genuinely different: it is
reachable from the release-publish hook, the feature-spotlight hook, and the
on-demand POST /video/request route, so two overlapping calls for the same
occasion can both pass the "no open task yet" check before either commits.
Fixed by wrapping the check+insert in a short-lived Redis mutex (reusing
HeartbeatMutex) keyed by occasion, mirroring XPostService's existing
lock pattern, with a regression test proving only one of two concurrent
calls creates a task.
Verified ReleaseExecutor's half-landed retry path (release_commit_sha):
apply_version_bumps and write_changelog_entry both run as uncommitted
working-tree edits before commit_and_push's single `git add -A` + commit,
so a bumped-version-without-changelog state can never reach origin (and
therefore can never be observed by a fresh retry clone) - confirmed correct
with a real-git-repo regression test, no fix needed.
Fixed sequencing.py's dev_task_collision_edges: the `if edges: return edges`
short-circuit dropped the same-assignee-lane fallback entirely whenever ANY
surfaced sibling pair produced a collision edge, even for a completely
unrelated same-assignee pair with no declared surface. Now the fallback
always runs, skipping only pairs the analyzer already ordered (so the two
mechanisms can never disagree on direction for the same pair).
Verified sequencing.py rule 3 (all-shared batch generates no edges): correct
by inspection (_shared_last_edges skips every pair when both are shared) and
confirmed with a regression test - no fix needed.
* [0c46f666] docs(reference): concurrency audit summary - engine races, fixes, verified patterns
---------
Co-authored-by: Backend Developer 2 <be-dev-2@roboco.tech>
Co-authored-by: Backend Documenter <be-doc@roboco.tech>
* [8f7f167a] Redis mutex pre-lock write audit (#629)
* [8f7f167a] Redis mutex pre-lock write audit: add cross-session regression test for XPostService.approve
Audited x_post_service.py, video_post_service.py, release_proposal.py, and
heartbeat_mutex.py for the pre-lock DB-write anti-pattern (a session write
that happens before the SET NX / HeartbeatMutex acquire returns a token,
letting a losing racer's stale write clobber a winner's committed state).
XPostService.approve, VideoPostService.approve, and
ReleaseProposalService.approve/reject already implement the correct
validate-pure-pre-lock, apply-under-lock pattern (the XPostService fix
already shipped per CHANGELOG.md: "X edited_body write deferred into the
single-flight lock (M5)"). HeartbeatMutex holds no AsyncSession at all, so
the anti-pattern is structurally inapplicable there.
Adds a genuine cross-session concurrency regression test to
test_x_post_service.py (a real second DB connection, not an in-process
mock) mirroring VideoPostService's existing cross-session test, proving a
concurrently-committed post survives and the CEO's edited body never lands
on the just-posted row.
* [8f7f167a] Remove redundant inline comments flagged by QA in cross-session regression test
Both comments restated what the surrounding docstrings already say
explicitly, per QA findings F-dbadd8f0 (line 294) and F-27ac051e (line
631) — no behavior change, tests re-verified green against a sandbox
Postgres.
* [8f7f167a] Remove inline trailing comments flagged by QA (correct file this time)
QA findings F-e6f3e6a6 and F-24189858 cited tests/unit/services/
test_x_post_service.py:294 and :631 across 5 revision rounds, but that
file never contained the flagged comment text — a repo-wide grep for
the exact quoted strings shows both comments actually live in the
mirrored tests/unit/services/test_video_post_service.py file, in its
own cross-session concurrency regression tests (the caption-edit and
tiktok-skip tests). Removed both there:
- "# externally visible to the "concurrent" session below" on the
db_session.commit() call
- "# never attempted without credentials" on the tiktok_poster.calls
assertion
Both restated what the surrounding docstrings/test names already say;
no behavior change. Verified with the full make quality gate against a
sandbox Postgres/Redis: 13,717 passed, 94.41% coverage, clean except
one pre-existing unrelated failure in tests/unit/api/test_cloud_auth.py
::test_login_route_parses_oauth2_form_not_query_params, which connects
to the app's default localhost:5432 Postgres (not the db_session
sandbox fixture) and is unreachable in this sandboxed environment —
structurally unrelated to the auth subsystem this task never touches.
* [8f7f167a] Redis mutex pre-lock write audit (round 7): add cross-session regression tests for reject() lock protection
Round-7 QA findings F-7eb9fbcb, F-06f39a2e, and F-4d56e49b claim
XPostService.reject(), ReleaseProposalService.reject(), and
release_executor._await_proc() lack lock protection / a CancelledError
handler — but their cited line ranges (255-267, 429-454, 241-257)
describe a pre-fix, shorter version of these functions that predates
commit
|
||
|
Renn F
|
a1233b2aeb |
fix(panel): V6 review gaps — honest errors, safe secretary start, real tests
CEO A2A mutations now invalidate the Mine-list query key so the list refreshes without the socket; the tg Metrics tab renders explicit error notes instead of confident zero stats when a section's fetch fails; the tg Secretary chat checks a new registry-backed /secretary/live/active route before auto-starting, showing a Take Over button instead of silently killing a session live on another device; the AI-routing card surfaces roster fetch errors instead of an empty grid; the shared acceptance-criteria editor caps at the backend's 7-item limit; the board-tab comment no longer calls the task sheet read-only. Tests: task-sheet approve/reject interactions (not just visibility), a non-demo metrics error-state test, secretary take-over branches, and dashboard-router auth-gate coverage (the e2e harness now mounts the dashboard router so the gate is actually exercised). |
||
|
Renn F
|
5ca8a9c4a6 |
fix(github-app): PAT fallback covers every mint failure; video/motion/x cleanups
mint_installation_token now wraps JWT build + HTTP + parsing so raw httpx/jwt failures surface as GitHubAppError and the existing PAT fallback catches them all (a GitHub outage no longer crashes git operations for App-bound projects). The PEM is validated at credential-set time instead of first mint, and the installation-token cache is cleared when credentials are deleted. Also: the video preview root resolves symlinks like its sibling route (frames under a symlinked workspaces_root no longer false-404), the tracked dangling motion/node_modules symlink is removed and the gitignore gains a slash-less entry that actually matches symlinks, changelog caption input strips GHSA refs like PR refs, and the edit-project dialog hides the GitHub App section for auto-detected gitlab.com projects and warns before a save that would clear both auth sources. |
||
|
Renn F
|
0296ec6fde |
fix(settings): ceo_name write-through renames the CEO agent row
The ceo_name setting only reached the panel header — the CEO agent row kept its seeded name, so agent selectors (create-flag, assignee pickers, kanban) still showed the old name after a rename. Setting ceo_name now updates the CEO row's name in the same transaction, resolved via the shared duplicate-tolerant get_agent_by_role lookup. The seed path never overwrites existing rows, so a custom name survives restarts. The agent-facing docs/map entries drop the personal name. Write-through tests pin ancient created_at on fixture rows so they stay deterministic against CEO rows other suite tests may have committed. |
||
|
Renn F
|
5143aab698 |
fix(notifications): one duplicate-tolerant role lookup for every singleton-role site
The duplicate-CEO-row fix landed on one call site but three identical bare 'select(...).where(role == ...)' + scalar_one_or_none() lookups remained in the orchestrator (auditor + two CEO), each still raising MultipleResultsFound under the same condition. All five sites now route through a shared get_agent_by_role helper (earliest-created row wins). The five event-bus notification handlers also thread task_title so a revived producer renders titles instead of #id8 (protocol signatures updated to match the service), the pull_request CI trigger mirrors the push trigger's paths so a fork PR touching only those paths still runs CI, and a stale a2a comment about auditor/pr_reviewer lacking read_a2a is corrected. |
||
|
Renn F
|
161b36b563 |
fix(git): stage only codegen-produced drift and link the auto-commit
The pre-push codegen auto-commit staged with a blanket 'git add -A', so any pre-existing dirty file in the worktree (e.g. crash-orphaned edits a resume never cleans) was silently swept into the 'regenerate generated artifacts' commit. Now the porcelain state is snapshotted before and after the codegen run and only newly-dirty paths are staged with 'git add --'. The commit is also recorded via _link_commit_to_task with the pushing agent's id threaded from both push call sites, so it shows up in the task's commit history like every other commit. |
||
|
Renn F
|
07bdef3db2 |
fix(guard): scope the internal-mesh whitelist to loopback + docker bridge
The WAF/IP-ban/rate-limit exemption whitelisted all of RFC1918, so any LAN client reaching the host-published nginx port resolved to its real 192.168.x.x and rode the exemption. Only the docker agent mesh needs it: keep 127.0.0.1/::1 + 172.16.0.0/12, drop 10.0.0.0/8 and 192.168.0.0/16. New middleware tests drive the real XFF resolution path (forwarded LAN client blocked, direct bridge peer exempt); the remaining ceiling (host-proxied traffic resolving to loopback/bridge-gateway under depth-1 XFF trust) is documented instead of claimed away. |
||
|
|
34a4950918 |
fix(panel): A2A transcript/list poll as fallback when the socket drops (#639)
* fix(panel): A2A transcript/list poll as fallback when the socket drops The desktop A2A view refreshed ONLY on /ws/system a2a.message frames — refetchInterval was off (only the /tg mini app polled). So when the socket flaps (the NAS stack flaps often), the open transcript froze: agent replies never landed and the thread stuck on the last frame received, even though the messages persisted server-side. useA2AMessages and useA2AConversations now take a 10s REST poll gated on the live-stream connection — polls only while disconnected, never when the WS is healthy, so it's a true fallback with no wasted requests. Backend was fine (get_messages_admin returns the full transcript incl. CEO interjects; the frame carries the right conversation_id). * fix(panel): make the A2A poll an unconditional backstop, not disconnect-gated Live-checked the NAS: /ws/system connections stay open (10 opens / 0 closes in 30m) and events publish — the socket is NOT flapping, so a disconnect-gated poll wouldn't fire. The real freeze mode is a silent-dead / half-open WS that still reports readyState OPEN (no client keepalive ping detects it), where isConnected stays true. So poll unconditionally: 20s while the socket claims up, 8s once known-down. Guarantees liveness regardless of why a frame didn't land. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
c83482ad19 |
feat(panel): bind an existing project to the GitHub App from Edit (#633)
#621 only let a NEW project bind to a GitHub App installation (the create dialog's Select repo picker). An already-imported project on a PAT had no way to re-route to the App. The Edit Project dialog now carries a GitHub App section: when App creds are configured, it shows the current binding (App installation vs PAT), reuses the same SelectRepoPicker to bind, and an Unbind button to revert to PAT (sends explicit null). Hidden/disabled for non-GitHub providers. Once bound, git ops (commits, PR reviews) are attributed to the App bot instead of the operator's account. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
5f42a93b4f |
feat(git): auto-regenerate + commit codegen drift before push (#632)
A project that checks in generated artifacts (RoboCo's lifecycle renders, verb tables) drifts whenever their source changes. The agent pre-submit gate (make gate) omits foundation-check, so drift is invisible at the desk and only fails on CI's drift gate — a failure with no link back to the task, which made one live task thrash 8 revision rounds. New per-project codegen_command (migration 078): run in the task's worktree right before push, and any drift committed into the same push, so CI never sees stale artifacts. Fail-open — a broken/timeout codegen command logs and lets the push proceed (CI's drift gate is the safety net); a null command (every project without checked-in codegen) is a pure no-op. Hooked at both push_branch (open_pr's first push, the PR head CI grades) and push_task_branch (later re-pushes). RoboCo sets codegen_command='make codegen' (a new Makefile target — the write counterpart to foundation-check's read) via the panel. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
0527e9ebf3 |
feat(panel): Secretary/Intake cards get a chat icon that opens their screen (#624)
The Auditor and PR reviewers get the real DM button (this branch). Secretary and Intake aren't A2A-DMable — they run their conversation over a live-session bridge — so they now carry the same chat icon but it navigates to their own screen instead: Intake -> /prompter, Secretary -> /business?tab=secretary. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
73c05cfa5e |
feat(a2a): CEO can DM the Auditor and PR reviewers (#623)
* feat(a2a): CEO can DM the Auditor and PR reviewers
A mid-flight PR reviewer or Auditor that's stuck was unreachable — the CEO
had no way to DM them. Both roles now carry dm/read_a2a, so the CEO can open
a 1:1 and they can reply in-thread through the existing CEO-reply path.
Scoped deliberately: the Auditor stays a silent observer to its peers — it
gains no peer-initiation surface (can_a2a_direct routes it through
_check_auditor_a2a, which refuses every initiation target; it can only reply
inside a CEO-opened DM). PR reviewers keep their owning-PM scope. Intake and
Secretary stay excluded — they have their own dedicated chat pages.
NO_COMMS_ROLES drops to {prompter, secretary}; the panel's EXCLUDE_NON_DM_ROLES
matches. KB/docs updated so the 'auditor/pr_reviewer have no dm' claim isn't
left stale.
* test(a2a): smoke guard checks _NO_COMMS_ROLES, not a hardcoded 'auditor'
The dm() runtime guard no longer names the auditor (it now carries dm to
reply to the CEO); it refuses the canonical _NO_COMMS_ROLES set. Assert on
that set so the smoke test tracks the guard, not a stale role name.
* chore(foundation): regenerate verb tables for auditor/pr_reviewer dm+read_a2a
---------
Co-authored-by: Renn F <rennf93@users.noreply.github.com>
|
||
|
|
775872cac0 |
fix(panel): days-view timeseries charts show dates, not '02:00' (#622)
formatBucket guessed hourly vs daily from the timestamp string
(bucket.endsWith('T00:00:00.000Z')) and then rendered LOCAL getHours(), so
every daily midnight-UTC bucket rendered as the viewer's local hour — '02:00'
at UTC+2 — across the 7d/30d/90d windows. Granularity is now derived from the
data's own bucket spacing (bucketGranularity: min gap >2h = daily) and daily
buckets render as a short UTC date. The 24h/hourly view is unchanged. Added
minTickGap so the 90d axis doesn't crowd.
Co-authored-by: Renn F <rennf93@users.noreply.github.com>
|
||
|
|
7b84162ae9 |
feat(github-app): App credentials, installation tokens, and a Select repo picker (#621)
* feat(github-app): App credentials, installation tokens, and a Select repo picker RoboCo was 100% PAT-based. A singleton Fernet-encrypted github_app_credentials row (migration 077, telegram-credentials pattern) now stores the App id + private key; github_app_auth mints RS256 app JWTs and caches installation tokens until 5 minutes before expiry. Projects can bind an installation (projects.github_installation_id): get_decrypted_token returns a minted installation token for bound projects and falls back to the stored PAT on any minting failure, so all ten token consumers work unchanged. CEO-gated routes expose credentials CRUD plus installation/repo listing, and the New Project dialog gains a Select repo picker (disabled with a HelpTip until the App is configured) that fills the git URL and binds the installation; manual URL + PAT stays the default path. * test(panel): mock the GitHub App credentials card in the settings page test --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
2d210ce6ee |
fix(notifications): CEO lookup tolerates duplicate rows; unbreak slave CI (#620)
#615 merged on a false green: the CI paths filter excludes motion/**, so a motion-only commit fired no quality run and the merge landed two latent breakages on slave. - _get_ceo_agent used scalar_one_or_none on role==CEO, which raises MultipleResultsFound once a second CEO-role row exists. It now pins to the earliest-created CEO, mirroring the sibling _get_auditor_agent. This is what made test_brand_voice_nudge_fires_once fail under the full-suite ordering. - _process_mentions tipped to xenon rank C when the skip-drafting branch was added; extracted the cap check and the skip filter into two small helpers. - ci.yml push paths now include motion/** so a motion-only commit can't false-green the quality gate again. Regression test: two CEO rows no longer break the lookup. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
16fa018ace |
feat(x): real voice guide, slop ban, and caption craft for drafted content (#615)
* feat(x): real voice guide, slop ban, and caption craft for drafted content Release posts and mention replies were drafted from a one-sentence voice stub while the reasoning-backed Head-of-Marketing voice guide only reached the off-by-default spotlight path. The drafting prompts now carry the full voice rules, a banned AI-slop list, three style exemplars, hook-first structure, and an under-240-char budget so the 280 clamp never truncates mid-sentence. An empty brand_voice now nudges the CEO exactly once (durable system_settings marker) instead of silently shipping baseline voice forever. A failed reply draft skips origination instead of shipping 'Thanks for the mention!'. Video dev prompts and motion/README gain per-platform caption templates (X: hook + specifics + outro; TikTok: hook + short lines + few niche hashtags). * docs(motion): reflow the new Captions section to satisfy the prose gate --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
c8f55be904 |
fix(notifications): task titles and agent slugs replace raw UUIDs (#616)
* fix(notifications): task titles and agent slugs replace raw UUIDs
Notification producers interpolated raw task/agent UUIDs into subjects and
bodies ('Task 68e1e4db-... unblocked', 'handed back to 00000000-...-0004').
A tiny notification_text helper (task_display: title-first with a #id8
fallback; agent_display: identity-map slug first, DB lookup fallback) now
feeds every producer: all 13 NotificationService methods, the 7
delivery-service bodies whose subjects were already title-based, the
substitute-PM ad-hoc insert, and the orchestrator/choreographer callers,
which thread the task row's title one call deeper. Fixes the literal
'cell_pm' role string sent as an agent slug in the merge-conflict
notification. Tool-call examples like unblock('<uuid>') keep the raw id on
purpose — agents need it.
* test(notifications): board-review subject assertion matches the humanized format
---------
Co-authored-by: Renn F <rennf93@users.noreply.github.com>
|
||
|
|
f6cca66afa |
fix(git): branch list classifies remote refs correctly and prunes stale ones (#610)
* fix(git): branch list classifies remote refs correctly and prunes stale ones The branches route detected remote-tracking refs via a 'remotes/' prefix that --format=%(refname:short) never emits, so every origin/* ref rendered under LOCAL and origin/HEAD surfaced as a fake branch. Listing now uses the full %(refname) and classifies on refs/heads/ vs refs/remotes/. Cleanup's remote deletion worked, but no code path ever pruned the viewing clone's remote-tracking refs, so deleted branches persisted in the UI forever. The branches route now runs a best-effort 'git remote prune origin' before listing remote refs, and the manual Fetch fetches with --prune. * refactor(git): extract branch-line classifier to satisfy the complexity gate --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
e125ef08aa |
feat(settings): CEO display name is configurable, Renzo hardcode removed (#612)
The header chip and the Settings User Info card rendered a literal 'Renzo'. The name now lives in the system_settings store under ceo_name (validated: trimmed, non-empty, max 60 chars) with the same client-served default the transcript-retention card uses, editable inline from the User Info card. Agent prompts already refer to 'the CEO' generically, so no prompt rewiring; the two agent-facing RAG docs drop the name too. License/CLA copyright is untouched. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
8cb233c1e8 |
fix(panel): forms catch up with the backend (#614)
New Project claimed GitLab/Gitea were 'planned' while both providers are fully shipped, showed a hardcoded GitHub badge, and never sent git_provider at all — a non-GitHub project could not be created without an immediate edit. It now carries the same forge Select the edit dialog has; the edit dialog's own 'GitLab support is planned' tooltip is corrected too. Also: the git actions panel's hardcoded 'main' (wrong PR-eligibility and target label for master-default and env-ladder projects) is replaced by the project's resolved head branch; acceptance criteria become editable in the edit-task dialog; three feature flags get their missing descriptions; and three forms swap raw-UUID text inputs for the existing Task/Agent selectors. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
fbec679878 |
fix(panel): AI Providers mix grid derives from the live roster (#613)
The per-agent override grid rendered a hand-maintained AGENT_GROUPS literal that had drifted: ux-dev-2 and all four PR reviewers were absent, so their model overrides could not be viewed or edited at all (the backend accepts any slug). The grid now derives its sections from useAgentDefinitions() with the same team helpers the Fleet page uses, org-ranked ordering, a loading skeleton, and group HelpTips. The static offline-fallback maps (agent-utils, use-agents, mock-data) get the missing agents too. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
7248e5b722 |
fix(panel): charts get real axes, humanized ticks, and dark-theme tooltips (#611)
Every chart hand-rolled its own k-only formatter (22596k-style ticks, left- clipped y labels), used recharts' default white tooltip (invisible header on the dark theme), and two charts pinned a numeric XAxis interval that collapses short series to a single tick. The agent Token Activity chart had no axes at all and blanked its tooltip date on purpose. One shared formatTokens/formatBucket (lib/format.ts) and one shared themed tooltip style (components/charts/chart-tooltip.tsx) now feed all 10 charts; axis widths/margins sized to the labels; preserveStartEnd tick intervals. Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
3c5ee46347 |
feat(tg): Mini App V6 — premium overhaul (#609)
* feat(tg): Mini App V6 — premium overhaul (design system, Chat parity, Metrics drilldown, CEO verbs) Design system: native type with tabular-numeral heroes (mono demoted to the wordmark), borderless elevated cards, floating dock, Telegram window-chrome painting via the theme bridge; Inbox moves behind a header bell with humanized notifications (UUIDs resolve to task names). Chat: honest Mine/Fleet split — participant-scoped CEO threads with real unread counts and mark-read, watched fleet threads with reply-as-CEO on task-linked conversations (watch-only otherwise), markdown transcripts, live pulse flashes, and a pinned Secretary live chat on the panel's SSE session runtime. Metrics: new tab with period-segmented spend hero, by-agent/team/model breakdowns, delivery + efficiency health, and a per-agent drilldown over usage time-series (agent_slug) + member scorecard. Board: tg-native grouped pipeline replacing the MobileTaskBoard wrapper; task sheet gains the CEO decide verbs (approve / request changes / unblock). Security: /api/dashboard router now require_panel_token-gated at router level (mirrors /api/usage), closing unauthenticated metrics exposure. * fix(tg): restore Share Tech Mono brand voice, Phosphor icon set, borderless avatars The mono returns as the numeral/brand voice (.tg-display — heroes, stat values, wordmark) while labels stay native sentence case. The hand-drawn duotone glyphs and lucide feature icons are replaced by Phosphor (MIT): duotone at rest via an IconContext at the shell, filled weight on the dock's active tab; row glyph maps (board statuses, inbox kinds, approval kinds, quick actions) all move over. Team avatar tiles drop their borders — tint-only squircles. * fix(tg): fleet avatar strip breathes — spaced tiles instead of overlap * polish(tg): taste-skill audit pass — em-dash purge, one icon family, separator rationing Applied the design-taste audit against the cockpit: every em-dash in visible UI copy rewritten (periods/commas/colons), the remaining lucide chrome (carets, arrows, send, close, spinners) moved to Phosphor so the tg tree ships one icon family (send is the native paper-plane, carets bold), the hand-rolled chevron SVG deleted, and metadata lines rationed to a single middle-dot separator. * polish(tg): pipeline chip strip scrolls without a visible scrollbar * fix(tests): metrics observability fixture uses a relative timestamp The hardcoded _T0 (2026-06-20) aged out of the service's 30-day window exactly 30 days later, detonating the suite on every branch. Two days back from now() stays inside every window (30d metrics, 7d scorecards) permanently. --------- Co-authored-by: Renn F <rennf93@users.noreply.github.com> |
||
|
|
cd73ad6a74 |
[68e1e4db] Video: release 0.26.0 (#604)
* [68e1e4db] feat(video): author release-0.26.0 marketing clip in the panel-demo kit register * [68e1e4db] fix(video): extend pk-frame clip window to full length to kill black-gap render bug * [68e1e4db] docs(video): document release-0.26.0 composition structure in motion/README Added a comprehensive "Release-specific example: release-0.26.0" section documenting the v0.26.0 marketing video composition: a 40-second panel-demo kit clip showcasing four key features (orchestrator security hardening, three-forge support, Telegram Mini App V4, active guard mode) with choreographed panel elements, stats overlay, and dual-format captions. Includes preview/test instructions, props.js shape, captions schema with verified character counts, and smoke-test invariants matching the pattern established by release-0.25.0. --------- Co-authored-by: UX/UI Developer 1 <ux-dev-1@roboco.tech> Co-authored-by: UX/UI Documenter <ux-doc@roboco.tech> |
||
|
|
a5d8c6bd5b |
feat(video): CEO can preview a video authoring task's frames before approving (#608)
A source=video authoring task reaches awaiting_ceo_approval with no MP4
yet — rendering only happens after it completes — so the CEO had nothing
to review. Two CEO-gated routes now serve the request_render preview
frames: GET /video/preview-frames/{task_id} lists them per orientation
(parsed from the self-describing .previews/{task8}/{orientation}/ filenames
rather than the render_preview marker, which only holds the last call's
single orientation), and .../{orientation}/{filename} streams a frame's
PNG behind the existing path-confinement guard. The task-detail Overview
gains a Video preview card — a 9:16/1:1 toggle + prev/next/scrubber frame
stepper with composition id, duration, and a dirty badge — shown for a
video task with preview frames or awaiting CEO approval.
Co-authored-by: Renn F <rennf93@users.noreply.github.com>
|