Merged
Conversation
What: replace the Claude-only shell flow with a Codex-first orchestrator and runner-enforced guard rails Fix: add a Python control plane, schema-backed cycle output, worktree verification, and updated install/docs paths
What: large-repo tmux testing showed repo-wide npm test inside Codex sandbox can fail even when runner verification passes Fix: instruct the agent to rely on runner baseline/final verification and only run narrow local checks when needed
What: large-repo tmux validation showed Codex could complete edits and targeted checks but still be killed before final JSON and commit bookkeeping Fix: make the runner timeout more forgiving so real repo cycles can finish while still staying bounded
fazxes
added a commit
that referenced
this pull request
Apr 5, 2026
All 3 looping daemons now: 1. Re-source lib-agent.sh every loop iteration (picks up new functions) 2. Self-restart via exec when daemon.sh changes (md5 hash comparison) Agent can modify shell scripts during a session and changes take effect next iteration automatically. Eliminates the #1 cause of lost autonomy.
This was referenced Apr 5, 2026
fazxes
added a commit
that referenced
this pull request
Apr 7, 2026
fazxes
added a commit
that referenced
this pull request
Apr 7, 2026
fazxes
added a commit
that referenced
this pull request
Apr 7, 2026
fazxes
added a commit
that referenced
this pull request
Apr 8, 2026
fazxes
added a commit
that referenced
this pull request
Apr 8, 2026
fazxes
added a commit
that referenced
this pull request
Apr 8, 2026
fazxes
added a commit
that referenced
this pull request
Apr 9, 2026
fazxes
added a commit
that referenced
this pull request
Apr 9, 2026
fazxes
added a commit
that referenced
this pull request
Apr 9, 2026
Quality audit findings and fixes (18 sessions since last audit): 1. OPERATIONS.md: test count 915 -> 1156 (stale by 241 tests) 2. OPERATIONS.md: add v0.0.7 and v0.0.8 version milestones (both missing) 3. OPERATIONS.md + CLAUDE.md: sync dependency flows -- both were divergent; add owl/eval_runner; correct settings/eval_targets ordering 4. DAEMON.md: arg 2 was "pause seconds" but daemon uses it as duration_hours 5. DAEMON.md: remove hardcoded absolute paths (/Users/no9labs/...) with wrong .recursive placement; replace with relative paths 6. DAEMON.md: remove stale pentest log references (v1 artifact, files don't exist) 7. ROLE-SCORING.md: add pentest_framework_tasks and sessions_since_eval signals (both active in pick-role.py since sessions #109 and #124 respectively) 8. sessions/index.md: fix corrupted role field in session 20260409-020609 (shell injection artifact from regex extractor: .*'"$LOG_FILE"2>/d -> brain) Tasks created: - #249: Regenerate MODULE_MAP.md (stale since session #1, shows only 3 modules) - #250: Fix DAEMON.md cycle lifecycle git commands (shows checkout+clean-fd, actual daemon only does fetch+reset) make check: 1156 tests pass
5 tasks
fazxes
added a commit
that referenced
this pull request
Apr 9, 2026
…session #1) The module-map generator only scanned top-level nightshift/*.py files, missing all 5 subpackages (core, settings, owl, raven, infra) and 24 of 27 modules. Fixed _module_paths to scan known subpackage dirs, updated _parse_modules to use relative path keys (e.g. core/constants), and updated _internal_dependencies to map nightshift.core.constants -> core/constants for correct dep ordering. The regenerated MODULE_MAP.md now shows all 27 modules with accurate dependency order matching the flow documented in CLAUDE.md. Added 5 new tests covering subpackage discovery, module count, dependency ordering, display format, and tests/ exclusion.
fazxes
added a commit
that referenced
this pull request
Apr 9, 2026
feat: regenerate MODULE_MAP.md (stale since session #1)
fazxes
added a commit
that referenced
this pull request
Apr 9, 2026
fazxes
added a commit
that referenced
this pull request
Apr 9, 2026
fazxes
added a commit
that referenced
this pull request
Apr 9, 2026
fazxes
added a commit
that referenced
this pull request
Apr 9, 2026
fazxes
added a commit
that referenced
this pull request
Apr 9, 2026
fazxes
added a commit
that referenced
this pull request
Apr 9, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Testing
python3 -m py_compile nightshift.pybash -n run.shbash -n test.shbash -n install.sh./run.sh --dry-run./test.sh --dry-runReclip-Reposter-Hub copy(acceptedlog_onlyrun with passing runner verification)Notes
mainwas not pushed; this PR is opened fromfeat/codex-nightshift-runner