fix(curator): only mark agent-created for background-review sediment (#19621)

Tighten the provenance semantics added in #19618: skills a user asks a
foreground agent to write via skill_manage(create) now stay invisible to
the curator. Only skills the background self-improvement review fork
sediments through skill_manage get the created_by=agent marker.

- tools/skill_provenance.py — new ContextVar module mirroring the
  _approval_session_key pattern: set_current_write_origin / reset /
  get / is_background_review. Default origin is 'foreground'; the
  review fork sets 'background_review'.
- run_agent.py — run_conversation() binds the ContextVar from
  self._memory_write_origin at the top of each call. The review fork
  runs on its own thread (fresh context), so foreground and review
  contexts never cross-contaminate.
- tools/skill_manager_tool.py — skill_manage(action='create') now
  only calls mark_agent_created() when is_background_review(). All
  other cases (foreground create, patch, edit, write_file, delete)
  continue as before.
- tests: test_skill_provenance.py (6 tests covering the ContextVar
  surface), split test_full_create_via_dispatcher into foreground
  vs. review-fork variants, curator status tests now mark-first.

Why: the agent routinely edits existing user skills on the user's
behalf; those writes must never flip provenance. And when a user
explicitly asks the foreground agent to create a skill, that skill
belongs to the user. The curator should only be cleaning up after
its own autonomous sediment from the review nudge loop.
This commit is contained in:
Teknium 2026-05-04 02:42:16 -07:00 committed by GitHub
parent bff484a51b
commit 3c070f9f9d
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
6 changed files with 234 additions and 4 deletions

View file

@ -531,13 +531,41 @@ class TestSkillManageDispatcher:
assert result["success"] is False
def test_full_create_via_dispatcher(self, tmp_path):
"""Foreground create does NOT mark the skill as agent-created.
Skills created by user-directed foreground turns belong to the user;
only the background self-improvement review fork should mark its
own sediment as agent-created (so the curator can later consolidate
or prune it).
"""
with _skill_dir(tmp_path):
raw = skill_manage(action="create", name="test-skill", content=VALID_SKILL_CONTENT)
from tools.skill_usage import load_usage
usage = load_usage()
result = json.loads(raw)
assert result["success"] is True
assert usage["test-skill"]["created_by"] == "agent"
# No provenance marker on a foreground create — record either missing
# entirely (telemetry best-effort) or present with created_by unset.
rec = usage.get("test-skill") or {}
assert rec.get("created_by") in (None, "", False)
def test_create_from_background_review_marks_agent_created(self, tmp_path):
"""Background-review fork creates ARE marked as agent-created."""
from tools.skill_provenance import set_current_write_origin, BACKGROUND_REVIEW
token = set_current_write_origin(BACKGROUND_REVIEW)
try:
with _skill_dir(tmp_path):
raw = skill_manage(
action="create", name="review-sediment", content=VALID_SKILL_CONTENT
)
from tools.skill_usage import load_usage
usage = load_usage()
finally:
from tools.skill_provenance import reset_current_write_origin
reset_current_write_origin(token)
result = json.loads(raw)
assert result["success"] is True
assert usage["review-sediment"]["created_by"] == "agent"
def test_delete_via_dispatcher_threads_absorbed_into(self, tmp_path):
# Dispatcher must plumb absorbed_into through to _delete_skill so the