[napetrov]

Summary

Brings the journey catalog up to date with merged work and the PRs still in flight.

 J2  Make a skill
+    UC-J02-003 Improve a skill automatically (research, AC-J02-006..010 proposed)
 J3  + AC-J03-006 host-unsupported cells   + AC-J03-007 repo inspect   + AC-J03-008 repo budget
-J4  Check AGENTS.md or other instructions
+J4  Check AGENTS.md, other instructions or installed skills
     AC-J04-006 -> available, enforced;  + AC-J04-007..009 rev/patch arms, uptake
+    UC-J04-005 installed skill reaches the agent (AC-J04-010..012)
 J7  AC-J07-003 + host evidence;  + AC-J07-008 built-in review graders
-J8  Which models can handle this?
+J8  Which models or harnesses can handle this?
     + AC-J08-003 model provenance;  + UC-J08-002 harnesses, UC-J08-003 review agents (proposed)
 J10 AC-J10-008 -> partial;  + UC-J10-004..006 consistency, process metrics, multi-turn (proposed)
 J12 AC-J12-002 -> partial, enforced;  + AC-J12-003 CI task selection;  + AC-J12-004 (proposed)
 J14 AC-J14-005 -> enforced;  + AC-J14-008 unmeasured share;  + AC-J14-009..012 (proposed)
 J5/J6/J11  proposed criteria for #163, #137, #107, #176

• Each new proposed criterion has its open PR as decision_ref. The auto-improve ones have none yet, and their gap says "research".
• Enforced criteria are claimed by @pytest.mark.outcome markers on existing tests. No test logic changed.
• The repo inspect CLI tests moved to tests/test_repo_inspect_cli.py, so test_repo_environment.py stays under its growth baseline. The baseline is lowered from 1337 to 1305.

Evidence

• Before: 61 criteria, 24 enforced.
  After: 97 criteria, 38 enforced, 59 unverified. pytest --outcomes: 7802 passed, journeys verdict pass, no problems. mkdocs build --strict passes.

Merge Danger

Door: two-way

Docs, registry and test markers only.

Blast Radius: tests

A newly enforced criterion fails --outcomes if its marked tests are later removed or renamed.

🤖 Generated with Claude Code

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 92.01%. Comparing base (09be35b) to head (ddac56c).
❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #209   +/-   ##
=======================================
  Coverage   92.01%   92.01%           
=======================================
  Files         223      223           
  Lines       27768    27768           
  Branches     4563     4563           
=======================================
  Hits        25552    25552           
  Misses       1570     1570           
  Partials      646      646           

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.