mirror of
https://github.com/rennf93/roboco.git
synced 2026-08-03 07:23:24 +02:00
feat(content): obligate role note sections like journals
Completes the note(scope='handoff') write-path (WIP 23e6ee57): every role
with a dedicated note section is now obligated to populate it, the same way
journals are obligated.
Obligations (foundation.policy.tracing):
- DEV_NOTES / PR_REVIEWER_NOTES / QUICK_CONTEXT_MIN_CHARS requirements +
checkers, wired onto i_am_done (dev_notes), delegate (quick_context), and
pr_pass / pr_fail / post_pr_review (pr_reviewer_notes).
- Fixes a latent bug: the docs-notes checker read dev_notes instead of
doc_notes (the documenter's section); the i_documented shim now feeds
doc_notes to match.
Auditor: a session-scoped note obligation on i_am_idle — the auditor owns no
delivery task and has no delivery verb, so it must have recorded an
observation within the window before going idle (JournalService.has_recent_entry).
Write-then-gate: persisted sections (dev_notes / quick_context) are
pre-written by the agent's note(scope='handoff') before the gated verb;
argument-borne sections (doc_notes / pr_reviewer_notes) are checked through a
SimpleNamespace shim, the same pattern qa_notes already uses.
Config: dev/pr_reviewer/quick_context min-chars (40/40/30), panel-tunable.
Plus per-gap remediation hints and full coverage (write-path routing,
ownership, validation->remediation, each obligation, the doc_notes fix).
Full make-quality green: 9777 passed, 95.6% coverage.
This commit is contained in:
@@ -777,6 +777,21 @@ class Settings(BaseSettings):
|
||||
ge=1,
|
||||
description="Minimum characters for docs notes",
|
||||
)
|
||||
dev_notes_min_chars: int = Field(
|
||||
default=40,
|
||||
ge=1,
|
||||
description="Minimum characters for a developer's dev_notes section",
|
||||
)
|
||||
pr_reviewer_notes_min_chars: int = Field(
|
||||
default=40,
|
||||
ge=1,
|
||||
description="Minimum characters for a PR reviewer's pr_reviewer_notes section",
|
||||
)
|
||||
quick_context_min_chars: int = Field(
|
||||
default=30,
|
||||
ge=1,
|
||||
description="Minimum characters for a PM's quick_context resumption section",
|
||||
)
|
||||
|
||||
# Commit-validator thresholds (wired into the gateway commit() gate)
|
||||
commit_subject_min_chars: int = Field(
|
||||
|
||||
@@ -40,6 +40,15 @@ class Requirement(StrEnum):
|
||||
SELF_VERIFIED = "self_verified"
|
||||
NOTES_MIN_CHARS = "notes>=min"
|
||||
SUBTASKS_TERMINAL = "subtasks_terminal"
|
||||
# Role note-section obligations (parity with the journal requirements): a
|
||||
# role with a dedicated note section must populate it via
|
||||
# note(scope='handoff') the same way journals are obligated. The developer's
|
||||
# dev_notes, the PR reviewer's pr_reviewer_notes, and the PM's quick_context
|
||||
# had no agent write-path before — these obligate the section now that one
|
||||
# exists.
|
||||
DEV_NOTES_MIN_CHARS = "dev_notes>=min"
|
||||
PR_REVIEWER_NOTES_MIN_CHARS = "pr_reviewer_notes>=min"
|
||||
QUICK_CONTEXT_MIN_CHARS = "quick_context>=min"
|
||||
|
||||
|
||||
@dataclass(frozen=True)
|
||||
@@ -55,6 +64,9 @@ class GateContext:
|
||||
qa_notes_min_chars: int = 80
|
||||
docs_notes_min_chars: int = 20
|
||||
notes_min_chars: int = 20
|
||||
dev_notes_min_chars: int = 40
|
||||
pr_reviewer_notes_min_chars: int = 40
|
||||
quick_context_min_chars: int = 30
|
||||
|
||||
|
||||
@dataclass(frozen=True)
|
||||
@@ -149,10 +161,29 @@ def _check_qa_evidence_inspected(task: Any, _ctx: GateContext) -> list[str]:
|
||||
|
||||
|
||||
def _check_docs_notes_min_chars(task: Any, ctx: GateContext) -> list[str]:
|
||||
notes = getattr(task, "dev_notes", "") or ""
|
||||
notes = getattr(task, "doc_notes", "") or ""
|
||||
return [] if len(notes) >= ctx.docs_notes_min_chars else ["docs_notes>=min"]
|
||||
|
||||
|
||||
def _check_dev_notes_min_chars(task: Any, ctx: GateContext) -> list[str]:
|
||||
notes = getattr(task, "dev_notes", "") or ""
|
||||
return [] if len(notes) >= ctx.dev_notes_min_chars else ["dev_notes>=min"]
|
||||
|
||||
|
||||
def _check_pr_reviewer_notes_min_chars(task: Any, ctx: GateContext) -> list[str]:
|
||||
notes = getattr(task, "pr_reviewer_notes", "") or ""
|
||||
return (
|
||||
[]
|
||||
if len(notes) >= ctx.pr_reviewer_notes_min_chars
|
||||
else ["pr_reviewer_notes>=min"]
|
||||
)
|
||||
|
||||
|
||||
def _check_quick_context_min_chars(task: Any, ctx: GateContext) -> list[str]:
|
||||
notes = getattr(task, "quick_context", "") or ""
|
||||
return [] if len(notes) >= ctx.quick_context_min_chars else ["quick_context>=min"]
|
||||
|
||||
|
||||
def _check_docs_files_non_empty(task: Any, _ctx: GateContext) -> list[str]:
|
||||
docs = getattr(task, "documents", None) or []
|
||||
return [] if len(docs) >= 1 else ["docs_files_non_empty"]
|
||||
@@ -195,6 +226,9 @@ _CHECKERS: dict[Requirement, Checker] = {
|
||||
Requirement.SELF_VERIFIED: _check_self_verified,
|
||||
Requirement.NOTES_MIN_CHARS: _check_notes_min_chars,
|
||||
Requirement.SUBTASKS_TERMINAL: _check_subtasks_terminal,
|
||||
Requirement.DEV_NOTES_MIN_CHARS: _check_dev_notes_min_chars,
|
||||
Requirement.PR_REVIEWER_NOTES_MIN_CHARS: _check_pr_reviewer_notes_min_chars,
|
||||
Requirement.QUICK_CONTEXT_MIN_CHARS: _check_quick_context_min_chars,
|
||||
}
|
||||
|
||||
|
||||
@@ -232,8 +266,13 @@ VERB_REQUIREMENTS: dict[str, frozenset[Requirement]] = {
|
||||
Requirement.JOURNAL_DECISION_AT_CLAIM,
|
||||
}
|
||||
),
|
||||
# PM delegate — pre-gateway PM.md required journal:decision before each delegate.
|
||||
"delegate": frozenset({Requirement.JOURNAL_DECISION}),
|
||||
# PM delegate — pre-gateway PM.md required journal:decision before each
|
||||
# delegate. QUICK_CONTEXT_MIN_CHARS obligates the PM's resumption section
|
||||
# (quick_context) on the parent: satisfiable because the PM pre-writes it via
|
||||
# note(scope='handoff', section={done, next, ...}) before delegate.
|
||||
"delegate": frozenset(
|
||||
{Requirement.JOURNAL_DECISION, Requirement.QUICK_CONTEXT_MIN_CHARS}
|
||||
),
|
||||
# Developer submit — adds JOURNAL_DURING_WORK_AT_LEAST_ONE for mid-flight cadence.
|
||||
# SELF_VERIFIED is set by the auto-run in_progress→verifying transition; it
|
||||
# stays in the required-set as a defense-in-depth backstop.
|
||||
@@ -246,6 +285,10 @@ VERB_REQUIREMENTS: dict[str, frozenset[Requirement]] = {
|
||||
Requirement.JOURNAL_REFLECT,
|
||||
Requirement.JOURNAL_DURING_WORK_AT_LEAST_ONE,
|
||||
Requirement.ACCEPTANCE_CRITERIA_ADDRESSED,
|
||||
# The developer's dedicated section: obligated like the journal.
|
||||
# Satisfiable because the dev pre-writes dev_notes via
|
||||
# note(scope='handoff') before i_am_done (write-then-gate).
|
||||
Requirement.DEV_NOTES_MIN_CHARS,
|
||||
}
|
||||
),
|
||||
# QA pass/fail.
|
||||
@@ -263,12 +306,23 @@ VERB_REQUIREMENTS: dict[str, frozenset[Requirement]] = {
|
||||
Requirement.JOURNAL_LEARNING,
|
||||
}
|
||||
),
|
||||
# PR reviewer posts its change-request — must record a learning entry first.
|
||||
"post_pr_review": frozenset({Requirement.JOURNAL_LEARNING}),
|
||||
# PR reviewer posts its change-request — must record a learning entry first,
|
||||
# and fill its dedicated pr_reviewer_notes section. The section note is the
|
||||
# verb's own argument (review body), so it is checked via a SimpleNamespace
|
||||
# shim at the call site, not the persisted task (write-then-gate: the arg
|
||||
# isn't on the task yet — same pattern as qa_notes).
|
||||
"post_pr_review": frozenset(
|
||||
{Requirement.JOURNAL_LEARNING, Requirement.PR_REVIEWER_NOTES_MIN_CHARS}
|
||||
),
|
||||
# In-path PR-review gate: the reviewer records a learning entry before
|
||||
# passing or failing the assembled PR (parity with post_pr_review).
|
||||
"pr_pass": frozenset({Requirement.JOURNAL_LEARNING}),
|
||||
"pr_fail": frozenset({Requirement.JOURNAL_LEARNING}),
|
||||
# passing or failing the assembled PR (parity with post_pr_review), and
|
||||
# fills pr_reviewer_notes (the verb's notes/issues argument, shimmed).
|
||||
"pr_pass": frozenset(
|
||||
{Requirement.JOURNAL_LEARNING, Requirement.PR_REVIEWER_NOTES_MIN_CHARS}
|
||||
),
|
||||
"pr_fail": frozenset(
|
||||
{Requirement.JOURNAL_LEARNING, Requirement.PR_REVIEWER_NOTES_MIN_CHARS}
|
||||
),
|
||||
# Doc submit.
|
||||
"i_documented": frozenset(
|
||||
{
|
||||
|
||||
@@ -45,7 +45,10 @@ from roboco.services.gateway.remediation import (
|
||||
hint_for_missing_progress,
|
||||
hint_for_missing_qa_notes,
|
||||
hint_for_missing_reflect,
|
||||
hint_for_short_dev_notes,
|
||||
hint_for_short_doc_notes,
|
||||
hint_for_short_pr_reviewer_notes,
|
||||
hint_for_short_quick_context,
|
||||
hint_for_unaddressed_acceptance_criteria,
|
||||
)
|
||||
|
||||
@@ -1876,6 +1879,7 @@ class Choreographer:
|
||||
auto-verify path. It is filtered here and re-asserted by the spec
|
||||
action's own preconditions.
|
||||
"""
|
||||
from roboco.config import settings as _settings
|
||||
from roboco.foundation.policy import tracing as _tr
|
||||
|
||||
has_reflect = await self.journal.has_reflect_for_task(agent_id, task_id)
|
||||
@@ -1890,6 +1894,9 @@ class Choreographer:
|
||||
journal_learning_present=has_learning,
|
||||
journal_struggle_present=has_struggle,
|
||||
journal_during_work_count=during_work_count,
|
||||
# DEV_NOTES_MIN_CHARS reads the persisted task.dev_notes — the dev
|
||||
# pre-writes it via note(scope='handoff') before i_am_done.
|
||||
dev_notes_min_chars=_settings.dev_notes_min_chars,
|
||||
)
|
||||
requirements: list[_tr.Requirement] = [
|
||||
r
|
||||
@@ -2004,7 +2011,15 @@ class Choreographer:
|
||||
and (datetime.now(UTC) - latest).total_seconds() <= window_seconds
|
||||
)
|
||||
|
||||
ctx = _tr.GateContext(journal_decision_present=fresh)
|
||||
# QUICK_CONTEXT_MIN_CHARS applies only to ``delegate`` (its
|
||||
# required-set is the only one carrying it); the PM pre-writes the
|
||||
# parent's quick_context via note(scope='handoff') before delegating.
|
||||
# Setting the threshold here is inert for unblock / escalate, which do
|
||||
# not require it.
|
||||
ctx = _tr.GateContext(
|
||||
journal_decision_present=fresh,
|
||||
quick_context_min_chars=_settings.quick_context_min_chars,
|
||||
)
|
||||
result = _tr.check_requirements(
|
||||
task=t,
|
||||
requirements=list(_tr.requirements_for(verb)),
|
||||
@@ -2189,6 +2204,17 @@ class Choreographer:
|
||||
min_chars=_roboco_settings.docs_notes_min_chars
|
||||
),
|
||||
"docs_files_non_empty": hint_for_missing_doc_files(),
|
||||
"dev_notes>=min": hint_for_short_dev_notes(
|
||||
min_chars=getattr(_roboco_settings, "dev_notes_min_chars", 40),
|
||||
task_id=tid,
|
||||
),
|
||||
"pr_reviewer_notes>=min": hint_for_short_pr_reviewer_notes(
|
||||
min_chars=getattr(_roboco_settings, "pr_reviewer_notes_min_chars", 40),
|
||||
),
|
||||
"quick_context>=min": hint_for_short_quick_context(
|
||||
min_chars=getattr(_roboco_settings, "quick_context_min_chars", 30),
|
||||
task_id=tid,
|
||||
),
|
||||
"journal:note_at_claim": (
|
||||
"pre-gateway parity P1: write a journal:note at claim. "
|
||||
f"Call note(scope='note', task_id='{tid}', "
|
||||
@@ -2969,22 +2995,21 @@ class Choreographer:
|
||||
),
|
||||
context_briefing=briefing,
|
||||
)
|
||||
if guard := await self._pending_assignment_guard(agent_id, briefing):
|
||||
return await self._emit_rejection(
|
||||
guard, agent_id=agent_id, task_id=None, verb="i_am_idle"
|
||||
)
|
||||
if guard := await self._pm_unfinished_review_guard(agent_id, briefing):
|
||||
return await self._emit_rejection(
|
||||
guard, agent_id=agent_id, task_id=None, verb="i_am_idle"
|
||||
)
|
||||
if guard := await self._pm_uncovered_decomposition_guard(agent_id, briefing):
|
||||
return await self._emit_rejection(
|
||||
guard, agent_id=agent_id, task_id=None, verb="i_am_idle"
|
||||
)
|
||||
if guard := await self._pm_uncovered_required_cells_guard(agent_id, briefing):
|
||||
return await self._emit_rejection(
|
||||
guard, agent_id=agent_id, task_id=None, verb="i_am_idle"
|
||||
)
|
||||
# Pre-idle guards, evaluated in order — the first that returns an
|
||||
# Envelope short-circuits to a rejection (kept as a loop so adding a
|
||||
# guard doesn't push this verb over the return-count bound).
|
||||
idle_guards = (
|
||||
self._pending_assignment_guard,
|
||||
self._pm_unfinished_review_guard,
|
||||
self._pm_uncovered_decomposition_guard,
|
||||
self._pm_uncovered_required_cells_guard,
|
||||
self._auditor_note_guard,
|
||||
)
|
||||
for guard_fn in idle_guards:
|
||||
if guard := await guard_fn(agent_id, briefing):
|
||||
return await self._emit_rejection(
|
||||
guard, agent_id=agent_id, task_id=None, verb="i_am_idle"
|
||||
)
|
||||
paused_ids = await self._auto_pause_in_progress_tasks(agent_id)
|
||||
await self.task.mark_agent_idle(agent_id)
|
||||
if paused_ids:
|
||||
@@ -3108,6 +3133,47 @@ class Choreographer:
|
||||
context_briefing=briefing,
|
||||
)
|
||||
|
||||
# The auditor must have recorded an observation within this window before
|
||||
# it may go idle (session-scoped note obligation; see _auditor_note_guard).
|
||||
_AUDITOR_IDLE_NOTE_WINDOW_SECONDS: int = 3600
|
||||
|
||||
async def _auditor_note_guard(
|
||||
self, agent_id: UUID, briefing: dict[str, Any]
|
||||
) -> Envelope | None:
|
||||
"""Refuse i_am_idle when the auditor has not recorded an observation.
|
||||
|
||||
Every role with a dedicated note section is obligated to populate it,
|
||||
the same way journals are obligated. The auditor's section is
|
||||
``auditor_notes``, but it owns no delivery task and has no delivery
|
||||
verb to hang the obligation on — so the obligation is session-scoped:
|
||||
before going idle the auditor must have recorded an observation within
|
||||
the last ``_AUDITOR_IDLE_NOTE_WINDOW_SECONDS`` (a reflect journal entry,
|
||||
or a note(scope='handoff') section write — either leaves a journal
|
||||
entry). Inert for every other role.
|
||||
"""
|
||||
agent = await self.task.agent_for(agent_id)
|
||||
if agent is None or str(agent.role) != "auditor":
|
||||
return None
|
||||
if await self.journal.has_recent_entry(
|
||||
agent_id, self._AUDITOR_IDLE_NOTE_WINDOW_SECONDS
|
||||
):
|
||||
return None
|
||||
return Envelope.invalid_state(
|
||||
message=(
|
||||
"as the auditor you must record an observation before going "
|
||||
"idle — your auditor_notes section is the artifact the company "
|
||||
"reviews, so it cannot be left empty."
|
||||
),
|
||||
remediate=(
|
||||
"call note(scope='reflect', text='<what you observed and any "
|
||||
"concern>') for a journal observation, or "
|
||||
"note(scope='handoff', task_id='<task>', section={'summary': "
|
||||
"'<observation>', 'severity': 'info'|'watch'|'risk'}) to fill a "
|
||||
"task's auditor_notes — then retry i_am_idle()."
|
||||
),
|
||||
context_briefing=briefing,
|
||||
)
|
||||
|
||||
async def _pm_uncovered_decomposition_guard(
|
||||
self, agent_id: UUID, briefing: dict[str, Any]
|
||||
) -> Envelope | None:
|
||||
|
||||
@@ -272,13 +272,13 @@ class DocMixin(_Base):
|
||||
persisted to the task yet (the spec runner / verb body writes
|
||||
them via the atomic action / pre-dispatch stamp), so we thread
|
||||
them through a SimpleNamespace shim with the minimal attributes
|
||||
the foundation checkers read off the task object (dev_notes +
|
||||
the foundation checkers read off the task object (doc_notes +
|
||||
documents — see foundation.policy.tracing._check_docs_notes_min_chars
|
||||
and _check_docs_files_non_empty).
|
||||
"""
|
||||
has_reflect = await self.journal.has_reflect_for_task(doc_agent_id, task_id)
|
||||
task_view = SimpleNamespace(
|
||||
dev_notes=notes,
|
||||
doc_notes=notes,
|
||||
documents=list(files),
|
||||
)
|
||||
ctx = _tr.GateContext(
|
||||
|
||||
@@ -229,7 +229,9 @@ class PRGateMixin(_Base):
|
||||
if isinstance(pre, Envelope):
|
||||
return pre
|
||||
t, agent, role_str, briefing, spec_ctx = pre
|
||||
gate = await self._gate_tracing(reviewer_agent_id, task_id, t, role_str, verb)
|
||||
gate = await self._gate_tracing(
|
||||
reviewer_agent_id, task_id, t, role_str, verb, notes=notes
|
||||
)
|
||||
if gate is not None:
|
||||
return gate
|
||||
runner = self._verb_runner()
|
||||
@@ -332,14 +334,28 @@ class PRGateMixin(_Base):
|
||||
t: Any,
|
||||
role_str: str,
|
||||
verb: str,
|
||||
*,
|
||||
notes: str,
|
||||
) -> Envelope | None:
|
||||
"""pr_pass / pr_fail require a journal:learning entry (parity with QA)."""
|
||||
"""pr_pass / pr_fail require a journal:learning entry (parity with QA)
|
||||
plus a substantive pr_reviewer_notes section.
|
||||
|
||||
The section note is the verb's own ``notes`` argument (the review
|
||||
verdict / issues), not yet persisted to the task, so it is threaded
|
||||
through a SimpleNamespace shim (the foundation checker reads
|
||||
``task.pr_reviewer_notes`` — same write-then-gate pattern as qa_notes).
|
||||
"""
|
||||
from roboco.config import settings as _settings
|
||||
|
||||
has_learning = await self.journal.has_learning_for_task(
|
||||
reviewer_agent_id, task_id
|
||||
)
|
||||
ctx = _tr.GateContext(journal_learning_present=has_learning)
|
||||
ctx = _tr.GateContext(
|
||||
journal_learning_present=has_learning,
|
||||
pr_reviewer_notes_min_chars=_settings.pr_reviewer_notes_min_chars,
|
||||
)
|
||||
result = _tr.check_requirements(
|
||||
task=SimpleNamespace(),
|
||||
task=SimpleNamespace(pr_reviewer_notes=notes),
|
||||
requirements=list(_tr.requirements_for(verb)),
|
||||
ctx=ctx,
|
||||
)
|
||||
|
||||
@@ -293,7 +293,7 @@ class PRReviewerMixin(_Base):
|
||||
verb="post_pr_review",
|
||||
)
|
||||
gate = await self._pr_review_tracing_gate(
|
||||
reviewer_agent_id, task_id, t, role_str
|
||||
reviewer_agent_id, task_id, t, role_str, body=body
|
||||
)
|
||||
if gate is not None:
|
||||
return gate
|
||||
@@ -359,15 +359,33 @@ class PRReviewerMixin(_Base):
|
||||
}
|
||||
|
||||
async def _pr_review_tracing_gate(
|
||||
self, reviewer_agent_id: UUID, task_id: UUID, t: Any, role_str: str
|
||||
self,
|
||||
reviewer_agent_id: UUID,
|
||||
task_id: UUID,
|
||||
t: Any,
|
||||
role_str: str,
|
||||
*,
|
||||
body: str,
|
||||
) -> Envelope | None:
|
||||
"""post_pr_review requires a journal:learning entry (parity with QA)."""
|
||||
"""post_pr_review requires a journal:learning entry (parity with QA)
|
||||
plus a substantive pr_reviewer_notes section.
|
||||
|
||||
The section note is the verb's own ``body`` argument (the change
|
||||
request), not yet persisted to the task, so it is threaded through a
|
||||
SimpleNamespace shim (the foundation checker reads
|
||||
``task.pr_reviewer_notes`` — same write-then-gate pattern as qa_notes).
|
||||
"""
|
||||
from roboco.config import settings as _settings
|
||||
|
||||
has_learning = await self.journal.has_learning_for_task(
|
||||
reviewer_agent_id, task_id
|
||||
)
|
||||
ctx = _tr.GateContext(journal_learning_present=has_learning)
|
||||
ctx = _tr.GateContext(
|
||||
journal_learning_present=has_learning,
|
||||
pr_reviewer_notes_min_chars=_settings.pr_reviewer_notes_min_chars,
|
||||
)
|
||||
result = _tr.check_requirements(
|
||||
task=SimpleNamespace(),
|
||||
task=SimpleNamespace(pr_reviewer_notes=body),
|
||||
requirements=list(_tr.requirements_for("post_pr_review")),
|
||||
ctx=ctx,
|
||||
)
|
||||
|
||||
@@ -71,3 +71,29 @@ def hint_for_missing_doc_files() -> str:
|
||||
"i_documented(files=[...]) requires the list of doc-file paths you "
|
||||
"committed; pass at least one path"
|
||||
)
|
||||
|
||||
|
||||
def hint_for_short_dev_notes(*, min_chars: int, task_id: str) -> str:
|
||||
return (
|
||||
f"your dev_notes section is empty or under {min_chars} chars. Before "
|
||||
f"i_am_done, call note(scope='handoff', task_id='{task_id}', "
|
||||
"section={'summary': '<what you built, key changes, risks>'}) to fill "
|
||||
"it, then retry."
|
||||
)
|
||||
|
||||
|
||||
def hint_for_short_pr_reviewer_notes(*, min_chars: int) -> str:
|
||||
return (
|
||||
f"your review note must be at least {min_chars} chars stating what you "
|
||||
"checked and the verdict rationale; pass a longer `body` (post_pr_review)"
|
||||
" / `notes` (pr_pass) / `issues` (pr_fail)."
|
||||
)
|
||||
|
||||
|
||||
def hint_for_short_quick_context(*, min_chars: int, task_id: str) -> str:
|
||||
return (
|
||||
f"your quick_context section is empty or under {min_chars} chars. Before "
|
||||
f"delegate, call note(scope='handoff', task_id='{task_id}', "
|
||||
"section={'done': '<state so far>', 'next': '<what the cell should do>'})"
|
||||
" to leave a resumption handoff, then retry."
|
||||
)
|
||||
|
||||
@@ -6,7 +6,7 @@ Each agent has their own journal with entries tied to tasks and sessions.
|
||||
Integrates with the Optimal API for RAG indexing of entries.
|
||||
"""
|
||||
|
||||
from datetime import UTC, datetime
|
||||
from datetime import UTC, datetime, timedelta
|
||||
from typing import Any, ClassVar
|
||||
from uuid import UUID
|
||||
|
||||
@@ -861,6 +861,28 @@ class JournalService(BaseService):
|
||||
agent_id, task_id, JournalEntryType.STRUGGLE
|
||||
)
|
||||
|
||||
async def has_recent_entry(self, agent_id: UUID, within_seconds: int) -> bool:
|
||||
"""True iff the agent wrote any journal entry within the last
|
||||
``within_seconds`` (of any type, any task).
|
||||
|
||||
Backs the auditor's i_am_idle note obligation: the auditor has no
|
||||
delivery verb and no single owned task, so its section-fill obligation
|
||||
is session-scoped — it must have recorded an observation (a reflect
|
||||
journal entry, or a note(scope='handoff') section write, which also
|
||||
drops a journal trail entry) recently before going idle.
|
||||
"""
|
||||
cutoff = datetime.now(UTC) - timedelta(seconds=within_seconds)
|
||||
query = (
|
||||
select(func.count(JournalEntryTable.id))
|
||||
.join(JournalTable, JournalEntryTable.journal_id == JournalTable.id)
|
||||
.where(
|
||||
JournalTable.agent_id == agent_id,
|
||||
JournalEntryTable.timestamp >= cutoff,
|
||||
)
|
||||
)
|
||||
result = await self.session.execute(query)
|
||||
return (result.scalar() or 0) > 0
|
||||
|
||||
async def write_struggle(
|
||||
self,
|
||||
*,
|
||||
|
||||
@@ -31,6 +31,9 @@ def test_requirement_enum_has_canonical_values() -> None:
|
||||
"self_verified",
|
||||
"notes>=min",
|
||||
"subtasks_terminal",
|
||||
"dev_notes>=min",
|
||||
"pr_reviewer_notes>=min",
|
||||
"quick_context>=min",
|
||||
}
|
||||
actual = {r.value for r in tracing.Requirement}
|
||||
assert actual == expected, f"Requirement drift: {actual ^ expected}"
|
||||
@@ -52,6 +55,51 @@ def test_gate_result_has_passed_and_missing() -> None:
|
||||
assert result.missing == []
|
||||
|
||||
|
||||
def test_docs_notes_checker_reads_doc_notes_not_dev_notes() -> None:
|
||||
"""Regression: _check_docs_notes_min_chars must read ``doc_notes`` (the
|
||||
documenter's section), not ``dev_notes`` (the developer's)."""
|
||||
ctx = tracing.GateContext(docs_notes_min_chars=20)
|
||||
# A long dev_notes must NOT satisfy the docs requirement.
|
||||
only_dev = SimpleNamespace(dev_notes="x" * 50, doc_notes="")
|
||||
assert tracing._check_docs_notes_min_chars(only_dev, ctx) == ["docs_notes>=min"]
|
||||
# A long doc_notes satisfies it.
|
||||
has_doc = SimpleNamespace(dev_notes="", doc_notes="y" * 25)
|
||||
assert tracing._check_docs_notes_min_chars(has_doc, ctx) == []
|
||||
|
||||
|
||||
def test_note_section_checkers_read_their_own_fields() -> None:
|
||||
ctx = tracing.GateContext() # defaults: dev 40, pr_reviewer 40, quick_context 30
|
||||
assert (
|
||||
tracing._check_dev_notes_min_chars(SimpleNamespace(dev_notes="z" * 40), ctx)
|
||||
== []
|
||||
)
|
||||
assert tracing._check_dev_notes_min_chars(
|
||||
SimpleNamespace(dev_notes="z" * 39), ctx
|
||||
) == ["dev_notes>=min"]
|
||||
assert (
|
||||
tracing._check_pr_reviewer_notes_min_chars(
|
||||
SimpleNamespace(pr_reviewer_notes="z" * 40), ctx
|
||||
)
|
||||
== []
|
||||
)
|
||||
assert tracing._check_quick_context_min_chars(
|
||||
SimpleNamespace(quick_context="z" * 29), ctx
|
||||
) == ["quick_context>=min"]
|
||||
|
||||
|
||||
def test_note_section_obligations_wired_to_verbs() -> None:
|
||||
assert tracing.Requirement.DEV_NOTES_MIN_CHARS in tracing.requirements_for(
|
||||
"i_am_done"
|
||||
)
|
||||
assert tracing.Requirement.QUICK_CONTEXT_MIN_CHARS in tracing.requirements_for(
|
||||
"delegate"
|
||||
)
|
||||
for verb in ("pr_pass", "pr_fail", "post_pr_review"):
|
||||
assert tracing.Requirement.PR_REVIEWER_NOTES_MIN_CHARS in (
|
||||
tracing.requirements_for(verb)
|
||||
)
|
||||
|
||||
|
||||
def test_check_requirements_passes_when_all_satisfied() -> None:
|
||||
task = SimpleNamespace(
|
||||
plan={"x": 1},
|
||||
|
||||
@@ -424,6 +424,14 @@ async def test_dev_full_chain_through_awaiting_qa(
|
||||
assert refreshed is not None
|
||||
assert refreshed.pr_number == _PR_NUMBER, "PR recorded on task"
|
||||
|
||||
# i_am_done obligates the developer's dev_notes section — the agent fills
|
||||
# it via note(scope='handoff') first; record_section_note is that write.
|
||||
await task_service.record_section_note(
|
||||
task.id,
|
||||
"developer",
|
||||
{"summary": "Implemented /healthz and added a test for the happy path."},
|
||||
)
|
||||
|
||||
# 4. i_am_done — auto-runs in_progress → verifying → awaiting_qa.
|
||||
env = await c.i_am_done(dev_agent.id, task.id, "tests pass; route works")
|
||||
assert env.error is None, f"i_am_done failed: {env.message}"
|
||||
@@ -482,6 +490,12 @@ async def test_full_chain_through_doc_handoff(
|
||||
)
|
||||
await task_service.add_progress(task.id, dev_agent.id, "implemented /healthz")
|
||||
await c.open_pr(dev_agent.id, task.id)
|
||||
# i_am_done obligates the developer's dev_notes section (note(scope='handoff')).
|
||||
await task_service.record_section_note(
|
||||
task.id,
|
||||
"developer",
|
||||
{"summary": "Implemented /healthz and added a test for the happy path."},
|
||||
)
|
||||
env = await c.i_am_done(dev_agent.id, task.id, "tests pass; route works")
|
||||
assert env.error is None
|
||||
assert env.status == "awaiting_qa"
|
||||
|
||||
@@ -3,6 +3,7 @@
|
||||
from __future__ import annotations
|
||||
|
||||
import uuid
|
||||
from datetime import UTC, datetime, timedelta
|
||||
from typing import TYPE_CHECKING, Any
|
||||
from unittest.mock import AsyncMock as _AsyncMock
|
||||
from unittest.mock import MagicMock as _MagicMock
|
||||
@@ -11,7 +12,13 @@ from uuid import uuid4 as _u
|
||||
|
||||
import pytest
|
||||
import pytest_asyncio
|
||||
from roboco.db.tables import AgentTable, JournalTable, ProjectTable, TaskTable
|
||||
from roboco.db.tables import (
|
||||
AgentTable,
|
||||
JournalEntryTable,
|
||||
JournalTable,
|
||||
ProjectTable,
|
||||
TaskTable,
|
||||
)
|
||||
from roboco.models import AgentRole, AgentStatus, Team
|
||||
from roboco.models.base import (
|
||||
JournalEntryType,
|
||||
@@ -29,7 +36,7 @@ from roboco.models.journal import (
|
||||
TaskReflectionParams,
|
||||
)
|
||||
from roboco.services.journal import JournalService
|
||||
from sqlalchemy import select
|
||||
from sqlalchemy import select, update
|
||||
from sqlalchemy.exc import IntegrityError as _IE
|
||||
|
||||
if TYPE_CHECKING:
|
||||
@@ -730,3 +737,52 @@ async def test_search_entries_swallows_exception(
|
||||
|
||||
results = await svc.search_entries(aid, "query")
|
||||
assert results == []
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_has_recent_entry_false_then_true(journal_setup: dict) -> None:
|
||||
"""No entries → False; a freshly written entry is within the window."""
|
||||
svc: JournalService = journal_setup["svc"]
|
||||
agent_id = journal_setup["agent_id"]
|
||||
task_id = journal_setup["task_id"]
|
||||
|
||||
assert await svc.has_recent_entry(agent_id, 3600) is False
|
||||
|
||||
await svc.write_entry(
|
||||
agent_id=agent_id,
|
||||
title="observation",
|
||||
content="watching the seam between FE and BE",
|
||||
scope="reflect",
|
||||
task_id=task_id,
|
||||
)
|
||||
assert await svc.has_recent_entry(agent_id, 3600) is True
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_has_recent_entry_excludes_entries_outside_window(
|
||||
journal_setup: dict, db_session: AsyncSession
|
||||
) -> None:
|
||||
"""An entry older than the window does not count as recent."""
|
||||
svc: JournalService = journal_setup["svc"]
|
||||
agent_id = journal_setup["agent_id"]
|
||||
task_id = journal_setup["task_id"]
|
||||
|
||||
await svc.write_entry(
|
||||
agent_id=agent_id,
|
||||
title="stale observation",
|
||||
content="recorded two hours ago",
|
||||
scope="reflect",
|
||||
task_id=task_id,
|
||||
)
|
||||
# Backdate every entry on this agent's journal to two hours ago.
|
||||
journal = await svc.get_journal_by_agent(agent_id)
|
||||
assert journal is not None
|
||||
await db_session.execute(
|
||||
update(JournalEntryTable)
|
||||
.where(JournalEntryTable.journal_id == journal.id)
|
||||
.values(timestamp=datetime.now(UTC) - timedelta(hours=2))
|
||||
)
|
||||
await db_session.flush()
|
||||
|
||||
assert await svc.has_recent_entry(agent_id, 3600) is False
|
||||
assert await svc.has_recent_entry(agent_id, 3 * 3600) is True
|
||||
|
||||
@@ -437,6 +437,15 @@ async def test_dev_full_chain_through_awaiting_qa(
|
||||
env = await c.open_pr(dev_agent.id, task.id)
|
||||
assert env.error is None, f"open_pr failed: {env.message}"
|
||||
|
||||
# i_am_done now obligates the developer's dev_notes section — the agent
|
||||
# fills it via note(scope='handoff') first. record_section_note is the
|
||||
# service call that write-path makes.
|
||||
await task_service.record_section_note(
|
||||
task.id,
|
||||
"developer",
|
||||
{"summary": "Implemented /healthz and added a test for the happy path."},
|
||||
)
|
||||
|
||||
env = await c.i_am_done(dev_agent.id, task.id, "tests pass; route works")
|
||||
assert env.error is None, f"i_am_done failed: {env.message}"
|
||||
assert env.status == Status.AWAITING_QA.value
|
||||
|
||||
@@ -173,18 +173,14 @@ class _MockContentActions:
|
||||
*,
|
||||
agent_id: object,
|
||||
text: object,
|
||||
scope: str = "note",
|
||||
task_id: object = None,
|
||||
structured: object = None,
|
||||
**_kwargs: object,
|
||||
) -> Envelope:
|
||||
# `structured` mirrors the Wave 2 G4 production signature (panel
|
||||
# decision/reflect fields). The mock ignores it — the test asserts
|
||||
# lifecycle transitions, not journal-entry rendering.
|
||||
# The route forwards scope / task_id / structured / section as keyword
|
||||
# args; the stub absorbs them via **_kwargs (it asserts lifecycle
|
||||
# transitions, not journal-entry / section rendering). ``section``
|
||||
# mirrors the note(scope='handoff') write-path signature.
|
||||
_ = agent_id
|
||||
_ = text
|
||||
_ = scope
|
||||
_ = task_id
|
||||
_ = structured
|
||||
return Envelope.ok(status="noted", task_id=None, next="continue")
|
||||
|
||||
|
||||
|
||||
@@ -75,7 +75,8 @@ def _ready_task(task_id: Any, agent_id: Any) -> MagicMock:
|
||||
acceptance_criteria_status=[],
|
||||
commits=[{"sha": "deadbeef"}],
|
||||
documents=[],
|
||||
dev_notes="",
|
||||
# i_am_done obligates the developer's dev_notes section (>=40 chars).
|
||||
dev_notes="Implemented the change and added tests covering the new path.",
|
||||
quick_context=None,
|
||||
)
|
||||
|
||||
|
||||
@@ -0,0 +1,81 @@
|
||||
"""Tests for the auditor's i_am_idle note obligation (_auditor_note_guard).
|
||||
|
||||
Every role with a dedicated note section is obligated to populate it, like
|
||||
journals. The auditor owns no delivery task and has no delivery verb, so its
|
||||
obligation is session-scoped: it must have recorded an observation within the
|
||||
window before it may go idle. Inert for every other role.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
from typing import Any
|
||||
from unittest.mock import AsyncMock, MagicMock
|
||||
from uuid import uuid4
|
||||
|
||||
import pytest
|
||||
from roboco.services.gateway.choreographer import Choreographer, ChoreographerDeps
|
||||
|
||||
|
||||
def _make_deps(**overrides: Any) -> ChoreographerDeps:
|
||||
base: dict[str, Any] = {
|
||||
"task": AsyncMock(),
|
||||
"work_session": AsyncMock(),
|
||||
"git": AsyncMock(),
|
||||
"a2a": AsyncMock(),
|
||||
"journal": AsyncMock(),
|
||||
"audit": AsyncMock(),
|
||||
"evidence_repo": AsyncMock(),
|
||||
}
|
||||
base.update(overrides)
|
||||
return ChoreographerDeps(**base)
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_auditor_idle_blocked_without_recent_observation() -> None:
|
||||
"""An auditor with no recent journal entry is refused idle."""
|
||||
auditor_id = uuid4()
|
||||
task_svc = AsyncMock()
|
||||
task_svc.agent_for.return_value = MagicMock(role="auditor", team="board")
|
||||
journal = AsyncMock()
|
||||
journal.has_recent_entry.return_value = False
|
||||
c = Choreographer(_make_deps(task=task_svc, journal=journal))
|
||||
|
||||
guard = await c._auditor_note_guard(auditor_id, briefing={})
|
||||
|
||||
assert guard is not None
|
||||
body = guard.as_dict()
|
||||
assert body["error"] == "invalid_state"
|
||||
assert "auditor_notes" in body["message"]
|
||||
assert "note(scope='reflect'" in body["remediate"]
|
||||
# The window query was actually consulted.
|
||||
journal.has_recent_entry.assert_awaited_once()
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_auditor_idle_allowed_with_recent_observation() -> None:
|
||||
"""An auditor that recorded an observation recently may idle."""
|
||||
auditor_id = uuid4()
|
||||
task_svc = AsyncMock()
|
||||
task_svc.agent_for.return_value = MagicMock(role="auditor", team="board")
|
||||
journal = AsyncMock()
|
||||
journal.has_recent_entry.return_value = True
|
||||
c = Choreographer(_make_deps(task=task_svc, journal=journal))
|
||||
|
||||
guard = await c._auditor_note_guard(auditor_id, briefing={})
|
||||
|
||||
assert guard is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_idle_note_guard_inert_for_non_auditor() -> None:
|
||||
"""A developer never trips the auditor guard (no journal lookup at all)."""
|
||||
dev_id = uuid4()
|
||||
task_svc = AsyncMock()
|
||||
task_svc.agent_for.return_value = MagicMock(role="developer", team="backend")
|
||||
journal = AsyncMock()
|
||||
c = Choreographer(_make_deps(task=task_svc, journal=journal))
|
||||
|
||||
guard = await c._auditor_note_guard(dev_id, briefing={})
|
||||
|
||||
assert guard is None
|
||||
journal.has_recent_entry.assert_not_awaited()
|
||||
@@ -107,6 +107,7 @@ async def test_delegate_blocks_when_parent_assigned_to_other_agent() -> None:
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=other_pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
task_svc = AsyncMock()
|
||||
task_svc.get.return_value = parent
|
||||
@@ -131,6 +132,7 @@ async def test_delegate_allows_when_parent_in_progress_and_owned() -> None:
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
new_task = MagicMock(id=uuid4())
|
||||
task_svc = AsyncMock()
|
||||
@@ -158,6 +160,7 @@ async def test_delegate_blocks_when_subtask_cap_exceeded() -> None:
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
too_many = [MagicMock(id=uuid4()) for _ in range(13)]
|
||||
task_svc = AsyncMock()
|
||||
@@ -184,6 +187,7 @@ async def test_delegate_allows_when_subtask_cap_within_soft_zone() -> None:
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
many = [MagicMock(id=uuid4()) for _ in range(10)]
|
||||
new_task = MagicMock(id=uuid4())
|
||||
@@ -211,6 +215,7 @@ async def test_delegate_allows_at_zero_subtasks() -> None:
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
new_task = MagicMock(id=uuid4())
|
||||
task_svc = AsyncMock()
|
||||
@@ -236,6 +241,7 @@ async def test_delegate_blocks_at_exact_cap_plus_one() -> None:
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
# Already 12 children — adding the 13th must be blocked.
|
||||
twelve = [MagicMock(id=uuid4()) for _ in range(12)]
|
||||
@@ -250,3 +256,31 @@ async def test_delegate_blocks_at_exact_cap_plus_one() -> None:
|
||||
body = env.as_dict()
|
||||
assert body["error"] == "invalid_state"
|
||||
task_svc.create_subtask.assert_not_awaited()
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_delegate_blocks_when_parent_quick_context_empty() -> None:
|
||||
"""delegate obligates the PM's quick_context resumption section on the
|
||||
parent; an empty quick_context (PM never called note(scope='handoff'))
|
||||
fails the tracing gate before any subtask is created."""
|
||||
pm_id = uuid4()
|
||||
parent_id = uuid4()
|
||||
parent = MagicMock(
|
||||
id=parent_id,
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=pm_id,
|
||||
quick_context="",
|
||||
)
|
||||
task_svc = AsyncMock()
|
||||
task_svc.get.return_value = parent
|
||||
task_svc.agent_for.return_value = MagicMock(role="cell_pm", team="backend")
|
||||
task_svc.get_subtasks.return_value = []
|
||||
deps = _make_deps(task=task_svc)
|
||||
c = Choreographer(deps)
|
||||
|
||||
env = await c.delegate(pm_id, parent_id, _delegate_inputs())
|
||||
body = env.as_dict()
|
||||
assert body["error"] == "tracing_gap"
|
||||
assert "quick_context>=min" in body["missing"]
|
||||
task_svc.create_subtask.assert_not_awaited()
|
||||
|
||||
@@ -428,7 +428,7 @@ async def test_i_am_done_blocks_when_acceptance_criteria_unaddressed() -> None:
|
||||
pr_url="https://x/pr/8",
|
||||
team="backend",
|
||||
documents=[],
|
||||
dev_notes="",
|
||||
dev_notes="Implemented the change and added tests covering the new path.",
|
||||
)
|
||||
task_svc = AsyncMock()
|
||||
task_svc.get.return_value = t
|
||||
@@ -480,7 +480,7 @@ async def test_i_am_done_reflect_note_addresses_acceptance_criteria() -> None:
|
||||
pr_url="https://x/pr/8",
|
||||
team="backend",
|
||||
documents=[],
|
||||
dev_notes="",
|
||||
dev_notes="Implemented the change and added tests covering the new path.",
|
||||
qa_notes="",
|
||||
)
|
||||
task_svc = AsyncMock()
|
||||
@@ -534,7 +534,7 @@ async def test_i_am_done_blocks_when_journal_reflect_missing() -> None:
|
||||
pr_url="https://x/pr/8",
|
||||
team="backend",
|
||||
documents=[],
|
||||
dev_notes="",
|
||||
dev_notes="Implemented the change and added tests covering the new path.",
|
||||
)
|
||||
task_svc = AsyncMock()
|
||||
task_svc.get.return_value = t
|
||||
@@ -717,7 +717,7 @@ def _passing_i_am_done_task(agent_id: Any, task_id: Any) -> Any:
|
||||
pr_url="https://x/pr/8",
|
||||
team="backend",
|
||||
documents=[],
|
||||
dev_notes="",
|
||||
dev_notes="Implemented the change and added tests covering the new path.",
|
||||
qa_notes="",
|
||||
)
|
||||
|
||||
|
||||
@@ -618,6 +618,9 @@ async def test_delegate_parent_no_project_rejected() -> None:
|
||||
project_id=None,
|
||||
product_id=None,
|
||||
title="p",
|
||||
# delegate obligates the PM's quick_context; supply it so the
|
||||
# no-project guard is the load-bearing rejection.
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
task_svc = AsyncMock()
|
||||
task_svc.get.return_value = parent
|
||||
|
||||
@@ -672,6 +672,7 @@ async def test_delegate_main_pm_to_cell_pm_creates_subtask() -> None:
|
||||
project_id=project_id,
|
||||
status="in_progress",
|
||||
assigned_to=main_pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
new_task = MagicMock(id=uuid4())
|
||||
task_svc = AsyncMock()
|
||||
@@ -713,6 +714,7 @@ async def test_delegate_cell_pm_to_team_dev_creates_subtask() -> None:
|
||||
project_id=project_id,
|
||||
status="in_progress",
|
||||
assigned_to=cell_pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
new_task = MagicMock(id=uuid4())
|
||||
task_svc = AsyncMock()
|
||||
@@ -749,6 +751,7 @@ async def test_delegate_main_pm_to_dev_is_rejected() -> None:
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=main_pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
task_svc = AsyncMock()
|
||||
task_svc.get.return_value = parent
|
||||
@@ -784,6 +787,7 @@ async def test_delegate_cell_pm_to_other_pm_rejected() -> None:
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=cell_pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
task_svc = AsyncMock()
|
||||
task_svc.get.return_value = parent
|
||||
@@ -818,6 +822,7 @@ async def test_delegate_unknown_assignee_returns_invalid_state() -> None:
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
task_svc = AsyncMock()
|
||||
task_svc.get.return_value = parent
|
||||
@@ -852,6 +857,7 @@ async def test_delegate_invalid_team_enum_rejected() -> None:
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
task_svc = AsyncMock()
|
||||
task_svc.get.return_value = parent
|
||||
@@ -1128,6 +1134,7 @@ async def test_delegate_main_pm_to_cell_pm_rejects_code_typed_subtask() -> None:
|
||||
project_id=uuid4(),
|
||||
status="in_progress",
|
||||
assigned_to=main_pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
task_svc = AsyncMock()
|
||||
task_svc.get.return_value = parent
|
||||
@@ -1166,6 +1173,7 @@ async def test_delegate_main_pm_to_cell_pm_accepts_planning_subtask() -> None:
|
||||
project_id=project_id,
|
||||
status="in_progress",
|
||||
assigned_to=main_pm_id,
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
new_task = MagicMock(id=uuid4())
|
||||
task_svc = AsyncMock()
|
||||
|
||||
@@ -97,7 +97,8 @@ async def test_i_am_done_reassigns_task_to_qa_agent() -> None:
|
||||
# Gate Set E requires non-empty commits before submit_qa.
|
||||
commits=[{"sha": "abc"}],
|
||||
documents=[],
|
||||
dev_notes="",
|
||||
# i_am_done obligates the developer's dev_notes section (>=40 chars).
|
||||
dev_notes="Implemented the change and added tests covering the new path.",
|
||||
)
|
||||
after_verify = MagicMock(
|
||||
**{**initial.__dict__, "status": "verifying", "self_verified": True},
|
||||
@@ -162,7 +163,8 @@ async def test_i_am_done_skips_reassign_when_no_qa_agent() -> None:
|
||||
# Gate Set E requires non-empty commits before submit_qa.
|
||||
commits=[{"sha": "abc"}],
|
||||
documents=[],
|
||||
dev_notes="",
|
||||
# i_am_done obligates the developer's dev_notes section (>=40 chars).
|
||||
dev_notes="Implemented the change and added tests covering the new path.",
|
||||
)
|
||||
after_verify = MagicMock(
|
||||
**{**initial.__dict__, "status": "verifying", "self_verified": True},
|
||||
|
||||
@@ -92,7 +92,8 @@ def _ready_task(task_id: Any, agent_id: Any) -> MagicMock:
|
||||
],
|
||||
commits=[{"sha": "abc"}],
|
||||
documents=[],
|
||||
dev_notes="",
|
||||
# i_am_done obligates the developer's dev_notes section (>=40 chars).
|
||||
dev_notes="Implemented the change and added tests covering the new path.",
|
||||
)
|
||||
|
||||
|
||||
@@ -265,6 +266,41 @@ async def test_i_am_done_blocks_when_no_progress() -> None:
|
||||
task_svc.submit_qa.assert_not_awaited()
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Role note-section obligation: dev_notes must be filled (note(scope='handoff'))
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_i_am_done_blocks_when_dev_notes_empty() -> None:
|
||||
"""i_am_done obligates the developer's dev_notes section; an empty
|
||||
dev_notes (the dev never called note(scope='handoff')) fails the gate."""
|
||||
agent_id = uuid4()
|
||||
task_id = uuid4()
|
||||
t = _ready_task(task_id, agent_id)
|
||||
t.dev_notes = ""
|
||||
task_svc = AsyncMock()
|
||||
task_svc.get.return_value = t
|
||||
task_svc.agent_for.return_value = MagicMock(
|
||||
id=agent_id, role="developer", team="backend", slug=None
|
||||
)
|
||||
journal_svc = AsyncMock()
|
||||
journal_svc.has_reflect_for_task.return_value = True
|
||||
journal_svc.has_decision_for_task.return_value = True
|
||||
journal_svc.latest_decision_at.return_value = datetime.now(UTC)
|
||||
journal_svc.has_learning_for_task.return_value = False
|
||||
journal_svc.has_struggle_for_task.return_value = False
|
||||
deps = _make_deps(task=task_svc, journal=journal_svc)
|
||||
c = Choreographer(deps)
|
||||
|
||||
env = await c.i_am_done(agent_id, task_id, "done")
|
||||
body = env.as_dict()
|
||||
assert body["error"] == "tracing_gap"
|
||||
assert "dev_notes>=min" in body["missing"]
|
||||
assert "scope='handoff'" in body["remediate"]
|
||||
task_svc.submit_qa.assert_not_awaited()
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# E.5 happy path: all gates pass → submit_qa runs (NO catch-up)
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
@@ -62,6 +62,8 @@ def _parent_in_progress(pm_id: Any) -> MagicMock:
|
||||
status="in_progress",
|
||||
assigned_to=pm_id,
|
||||
priority=2,
|
||||
# delegate obligates the PM's quick_context resumption section.
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
|
||||
|
||||
|
||||
@@ -48,6 +48,8 @@ def _parent(pm_id: Any, product_id: Any = None, project_id: Any = None) -> Magic
|
||||
product_id=product_id,
|
||||
status="in_progress",
|
||||
assigned_to=pm_id,
|
||||
# delegate obligates the PM's quick_context resumption section.
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
|
||||
|
||||
|
||||
@@ -0,0 +1,167 @@
|
||||
"""Tests for note(scope='handoff') — the role note-section write-path.
|
||||
|
||||
``note()`` only ever wrote the JOURNAL; ``scope='handoff'`` is how an agent
|
||||
authors its dedicated SECTION (dev_notes / quick_context / auditor_notes …)
|
||||
through the ``apply_structured_note`` chokepoint. These cover the routing,
|
||||
ownership, validation→remediation, and journal-trail behaviours.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
from unittest.mock import AsyncMock, MagicMock
|
||||
from uuid import uuid4
|
||||
|
||||
import pytest
|
||||
from roboco.foundation.policy.content import ContentValidationError
|
||||
from roboco.services.gateway.content_actions import ContentActions, ContentActionsDeps
|
||||
|
||||
|
||||
def _make_deps(**overrides: object) -> ContentActionsDeps:
|
||||
task = overrides.get("task") or AsyncMock()
|
||||
return ContentActionsDeps(
|
||||
task=task,
|
||||
git=overrides.get("git") or AsyncMock(),
|
||||
messaging=overrides.get("messaging") or AsyncMock(),
|
||||
a2a=overrides.get("a2a") or AsyncMock(),
|
||||
journal=overrides.get("journal") or AsyncMock(),
|
||||
workspace=overrides.get("workspace") or AsyncMock(),
|
||||
notifications=overrides.get("notifications") or AsyncMock(),
|
||||
notification_delivery=overrides.get("notification_delivery") or AsyncMock(),
|
||||
evidence_repo=overrides.get("evidence_repo") or AsyncMock(),
|
||||
)
|
||||
|
||||
|
||||
def _dev_task_svc(task_id: object, role: str = "developer") -> AsyncMock:
|
||||
"""A task service whose active/context task is owned by the caller."""
|
||||
svc = AsyncMock()
|
||||
svc.agent_for.return_value = MagicMock(role=role)
|
||||
svc.get_journal_context_task_for_agent.return_value = MagicMock(id=task_id)
|
||||
svc.record_section_note.return_value = None
|
||||
return svc
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_handoff_developer_writes_dev_notes_from_text() -> None:
|
||||
"""A developer handoff routes to the 'developer' content type, defaulting
|
||||
the payload to {'summary': text} when no explicit section is given."""
|
||||
agent_id, task_id = uuid4(), uuid4()
|
||||
svc = _dev_task_svc(task_id)
|
||||
ca = ContentActions(_make_deps(task=svc))
|
||||
|
||||
summary = "Implemented the endpoint and added happy-path tests."
|
||||
env = await ca.note(agent_id=agent_id, text=summary, scope="handoff")
|
||||
|
||||
assert env.as_dict()["error"] is None
|
||||
svc.record_section_note.assert_awaited_once()
|
||||
called_task_id, content_type, payload = svc.record_section_note.call_args.args
|
||||
assert called_task_id == task_id
|
||||
assert content_type == "developer"
|
||||
assert payload == {"summary": summary}
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_handoff_passes_explicit_section_through() -> None:
|
||||
"""An explicit ``section`` dict is the payload (e.g. PM resumption)."""
|
||||
agent_id, task_id = uuid4(), uuid4()
|
||||
svc = _dev_task_svc(task_id, role="cell_pm")
|
||||
ca = ContentActions(_make_deps(task=svc))
|
||||
|
||||
section = {"done": "Planned the decomposition.", "next": "Cells implement."}
|
||||
env = await ca.note(
|
||||
agent_id=agent_id, text="handoff", scope="handoff", section=section
|
||||
)
|
||||
|
||||
assert env.as_dict()["error"] is None
|
||||
_tid, content_type, payload = svc.record_section_note.call_args.args
|
||||
assert content_type == "resumption"
|
||||
assert payload == section
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_handoff_also_writes_journal_trail_entry() -> None:
|
||||
"""The section write drops a journal trail entry so it shows in the log."""
|
||||
agent_id, task_id = uuid4(), uuid4()
|
||||
svc = _dev_task_svc(task_id)
|
||||
journal = AsyncMock()
|
||||
ca = ContentActions(_make_deps(task=svc, journal=journal))
|
||||
|
||||
await ca.note(agent_id=agent_id, text="Did the thing thoroughly.", scope="handoff")
|
||||
|
||||
journal.write_entry.assert_awaited_once()
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_handoff_role_without_section_is_rejected() -> None:
|
||||
"""A role with no dedicated section (board/advisory) cannot handoff."""
|
||||
agent_id = uuid4()
|
||||
svc = AsyncMock()
|
||||
svc.agent_for.return_value = MagicMock(role="product_owner")
|
||||
ca = ContentActions(_make_deps(task=svc))
|
||||
|
||||
env = await ca.note(agent_id=agent_id, text="observation", scope="handoff")
|
||||
body = env.as_dict()
|
||||
|
||||
assert body["error"] == "invalid_state"
|
||||
assert "no dedicated note section" in body["message"]
|
||||
svc.record_section_note.assert_not_awaited()
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_handoff_validation_error_returns_remediation_not_422() -> None:
|
||||
"""A malformed section payload becomes a remediation Envelope, never a 422
|
||||
(a raw 422 would trip the do-server circuit breaker)."""
|
||||
agent_id, task_id = uuid4(), uuid4()
|
||||
svc = _dev_task_svc(task_id, role="auditor")
|
||||
svc.record_section_note.side_effect = ContentValidationError(
|
||||
"severity", "field required"
|
||||
)
|
||||
ca = ContentActions(_make_deps(task=svc))
|
||||
|
||||
env = await ca.note(
|
||||
agent_id=agent_id,
|
||||
text="risk spotted",
|
||||
scope="handoff",
|
||||
section={"summary": "x"},
|
||||
)
|
||||
body = env.as_dict()
|
||||
|
||||
assert body["error"] == "invalid_state"
|
||||
assert "severity" in body["message"]
|
||||
assert "auditor" in body["remediate"]
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_handoff_no_task_to_attach_is_rejected() -> None:
|
||||
"""With no active/context task and no explicit task_id, handoff refuses."""
|
||||
agent_id = uuid4()
|
||||
svc = AsyncMock()
|
||||
svc.agent_for.return_value = MagicMock(role="developer")
|
||||
svc.get_journal_context_task_for_agent.return_value = None
|
||||
ca = ContentActions(_make_deps(task=svc))
|
||||
|
||||
env = await ca.note(agent_id=agent_id, text="orphan note", scope="handoff")
|
||||
body = env.as_dict()
|
||||
|
||||
assert body["error"] == "invalid_state"
|
||||
assert "task_id" in body["remediate"]
|
||||
svc.record_section_note.assert_not_awaited()
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_handoff_explicit_task_not_owned_is_rejected() -> None:
|
||||
"""An explicit task_id the caller does not own is an ownership violation."""
|
||||
agent_id, task_id = uuid4(), uuid4()
|
||||
svc = AsyncMock()
|
||||
svc.agent_for.return_value = MagicMock(role="developer")
|
||||
svc.get.return_value = MagicMock(
|
||||
id=task_id, assigned_to=uuid4(), project_id=uuid4(), product_id=None
|
||||
)
|
||||
ca = ContentActions(_make_deps(task=svc))
|
||||
|
||||
env = await ca.note(
|
||||
agent_id=agent_id, text="poking another task", scope="handoff", task_id=task_id
|
||||
)
|
||||
body = env.as_dict()
|
||||
|
||||
assert body["error"] == "not_authorized"
|
||||
svc.record_section_note.assert_not_awaited()
|
||||
@@ -89,10 +89,16 @@ def _make_deps(**overrides: Any) -> ChoreographerDeps:
|
||||
def _make_task(task_id: Any) -> Any:
|
||||
"""A task stub that the tracing gate accepts as-is.
|
||||
|
||||
`_check_pm_decision_required` only consults the (agent, task) journal
|
||||
lookup — the task object itself is opaque to that check.
|
||||
`_check_pm_decision_required` consults the (agent, task) journal lookup
|
||||
for the decision window and, for ``delegate``, the persisted
|
||||
``quick_context`` resumption section — so the stub carries a substantive
|
||||
quick_context to satisfy that obligation.
|
||||
"""
|
||||
return MagicMock(id=task_id, status="in_progress")
|
||||
return MagicMock(
|
||||
id=task_id,
|
||||
status="in_progress",
|
||||
quick_context="Decomposition planned; cells implement their slice next.",
|
||||
)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
@@ -0,0 +1,86 @@
|
||||
"""The PR reviewer's pr_reviewer_notes section is obligated on its verbs.
|
||||
|
||||
pr_pass / pr_fail (in-path gate) and post_pr_review (inbound external PR) each
|
||||
require a substantive review note. The note is the verb's own argument (not yet
|
||||
persisted), so it is checked through a SimpleNamespace shim against the
|
||||
``pr_reviewer_notes`` field — these cover both the short-circuit (too short) and
|
||||
the pass-through (long enough) at the tracing-gate helpers.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
from typing import Any
|
||||
from unittest.mock import AsyncMock, MagicMock
|
||||
from uuid import uuid4
|
||||
|
||||
import pytest
|
||||
from roboco.services.gateway.choreographer import Choreographer, ChoreographerDeps
|
||||
|
||||
|
||||
def _make_choreographer(*, has_learning: bool = True) -> Choreographer:
|
||||
base: dict[str, Any] = {
|
||||
"task": AsyncMock(),
|
||||
"work_session": AsyncMock(),
|
||||
"git": AsyncMock(),
|
||||
"a2a": AsyncMock(),
|
||||
"journal": AsyncMock(),
|
||||
"audit": AsyncMock(),
|
||||
"evidence_repo": AsyncMock(),
|
||||
}
|
||||
base["journal"].has_learning_for_task.return_value = has_learning
|
||||
return Choreographer(ChoreographerDeps(**base))
|
||||
|
||||
|
||||
_LONG = "Reviewed the assembled diff end to end; the seam contract holds."
|
||||
_SHORT = "looks ok"
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_gate_tracing_blocks_on_short_notes() -> None:
|
||||
c = _make_choreographer()
|
||||
env = await c._gate_tracing(
|
||||
uuid4(), uuid4(), MagicMock(), "pr_reviewer", "pr_pass", notes=_SHORT
|
||||
)
|
||||
assert env is not None
|
||||
body = env.as_dict()
|
||||
assert body["error"] == "tracing_gap"
|
||||
assert "pr_reviewer_notes>=min" in body["missing"]
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_gate_tracing_passes_on_substantive_notes() -> None:
|
||||
c = _make_choreographer()
|
||||
env = await c._gate_tracing(
|
||||
uuid4(), uuid4(), MagicMock(), "pr_reviewer", "pr_fail", notes=_LONG
|
||||
)
|
||||
assert env is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_post_pr_review_tracing_blocks_on_short_body() -> None:
|
||||
c = _make_choreographer()
|
||||
env = await c._pr_review_tracing_gate(
|
||||
uuid4(), uuid4(), MagicMock(), "pr_reviewer", body=_SHORT
|
||||
)
|
||||
assert env is not None
|
||||
assert "pr_reviewer_notes>=min" in env.as_dict()["missing"]
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_post_pr_review_tracing_passes_on_substantive_body() -> None:
|
||||
c = _make_choreographer()
|
||||
env = await c._pr_review_tracing_gate(
|
||||
uuid4(), uuid4(), MagicMock(), "pr_reviewer", body=_LONG
|
||||
)
|
||||
assert env is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_missing_learning_still_blocks_even_with_long_notes() -> None:
|
||||
"""The journal:learning requirement remains independent of the note."""
|
||||
c = _make_choreographer(has_learning=False)
|
||||
env = await c._gate_tracing(
|
||||
uuid4(), uuid4(), MagicMock(), "pr_reviewer", "pr_pass", notes=_LONG
|
||||
)
|
||||
assert env is not None
|
||||
assert "journal:learning" in env.as_dict()["missing"]
|
||||
@@ -6,7 +6,7 @@ from types import SimpleNamespace
|
||||
|
||||
import pytest
|
||||
from roboco.foundation.policy.content import ContentValidationError, PrReviewContent
|
||||
from roboco.services.content_notes import apply_structured_note
|
||||
from roboco.services.content_notes import apply_structured_note, content_type_for_role
|
||||
|
||||
|
||||
def _task() -> SimpleNamespace:
|
||||
@@ -95,3 +95,22 @@ def test_notes_structured_reassigned_for_dirty_tracking() -> None:
|
||||
assert t.notes_structured is not before # new dict object
|
||||
assert "developer" in t.notes_structured # prior entry preserved
|
||||
assert "doc" in t.notes_structured
|
||||
|
||||
|
||||
def test_content_type_for_role_maps_section_roles() -> None:
|
||||
"""Each role with a dedicated section routes to its content type."""
|
||||
assert content_type_for_role("developer") == "developer"
|
||||
assert content_type_for_role("qa") == "qa"
|
||||
assert content_type_for_role("documenter") == "doc"
|
||||
assert content_type_for_role("pr_reviewer") == "pr_review"
|
||||
assert content_type_for_role("auditor") == "auditor"
|
||||
assert content_type_for_role("cell_pm") == "resumption"
|
||||
assert content_type_for_role("main_pm") == "resumption"
|
||||
|
||||
|
||||
def test_content_type_for_role_none_for_sectionless_roles() -> None:
|
||||
"""Board / advisory / on-demand roles have no dedicated section."""
|
||||
assert content_type_for_role("product_owner") is None
|
||||
assert content_type_for_role("head_marketing") is None
|
||||
assert content_type_for_role("ceo") is None
|
||||
assert content_type_for_role("prompter") is None
|
||||
|
||||
Reference in New Issue
Block a user