BjornMelin/dev-skills / archive/skills/codex-session-cleanup
codex-session-cleanup
VerifiedSafely scan, triage, quarantine, restore, and purge Codex session/history/state/memory artifacts for current repo, explicit roots, subrepos, Codex-home, or all sessions.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 5.3 KB |
| agents/openai.yaml | config | 332 B |
| archive.json | config | 938 B |
| references/artifact-model.md | markdown | 2.5 KB |
| references/classifier-rules.md | markdown | 2.5 KB |
| references/report-template.md | markdown | 1.2 KB |
| references/safety-restore.md | markdown | 2.3 KB |
| scripts/codex_session_cleanup.py | script | 72.1 KB |
| tests/test_codex_session_cleanup.py | script | 22.5 KB |
references/safety-restore.md
View on GitHub# Safety And Restore
## Apply Safety
Before mutating, the script must have:
- a fresh scan manifest;
- exact `--confirm <manifest-id>`;
- `--execute`;
- readable and writable selected session files;
- writable quarantine path;
- backup inventory for existing core DB/history/index files;
- zero JSONL parse errors;
- passing SQLite integrity checks;
- no active goals, jobs, or unsafe spawn-edge references;
- no stale manifest or changed critical counts.
If any condition fails, stop.
## Memory Policy
Supported policies:
- `copy`: copy linked rollout summaries and a registry snapshot into quarantine
and leave active memory untouched. This is the default.
- `move`: move linked rollout summaries into quarantine after explicit
`--confirm-memory-move <manifest-id>`. The registry is still copied, not
moved.
- `ignore`: do not preserve linked memory summaries after explicit
`--confirm-memory-ignore <manifest-id>`.
If active developer or system instructions forbid memory mutation, use `copy`
even when a user previously preferred stronger cleanup.
## Autonomous Apply
Autonomous apply means reversible quarantine only. It is allowed only when:
- scope is narrow enough for reliable ownership;
- every selected thread is high-confidence;
- risk score is at or below the configured limit;
- age is at least the broad-scope threshold;
- memory policy is `copy`;
- no parser, backup, integrity, active-state, or durable-memory blockers exist;
- a read-only validator agrees when the report is broad or memory-linked.
Manual review is required for `--include-medium`, `--scope all`, destructive
memory policy, purge, and any disagreement or uncertainty.
## Restore
Restore with:
```bash
python3 scripts/codex_session_cleanup.py restore --quarantine <dir> --restore-db --execute
```
Default restore copies session and memory files back when missing. `--restore-db`
copies DB/history backups back over current files. Use it only when no newer
Codex state must be preserved.
## Purge
Purge is explicit and separate. First run without `--execute`; the command will
tell you the required `purge:<bundle-names>` confirmation token:
```bash
python3 scripts/codex_session_cleanup.py purge --older-than-days 30 --confirm purge:<bundle-names>
```
Do not purge a quarantine until rollback is no longer needed. Purge requires a
valid `result.json` and `SHA256SUMS` bundle.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |