@jy0xlu/codexgo
ARecover the previous actionable Codex request for the current workspace at the start of a fresh session after compaction, crash, or context loss. Use when the user types `codexgo` as an explicit skill request, asks to recover the previous Codex session, or asks to continue after a compact/crash break.
Install
agr install @jy0xlu/codexgo --target claudeWrites 5 files into .claude/skills/, pinned to git-dab583fc.
- .claude/skills/codexgo/.gitignore
- .claude/skills/codexgo/LICENSE
- .claude/skills/codexgo/README.en.md
- .claude/skills/codexgo/README.md
- .claude/skills/codexgo/SKILL.md
Document
name: "codexgo"
description: "Recover the previous actionable Codex request for the current workspace at the start of a fresh session after compaction, crash, or context loss. Use when the user types codexgo as an explicit skill request, asks to recover the previous Codex session, or asks to continue after a compact/crash break."
codexgo
Use this skill only when the user explicitly wants to recover the previous Codex session for the current workspace, usually at the start of a fresh session after compaction or crash.
Do not auto-trigger this skill later in a normal active conversation from vague phrases like "continue", "继续", or "接着做". In an active thread, prefer the current conversation context unless the user clearly names this skill.
Run
Run:
python scripts/codexgo.py --cwd "$PWD" --format json
On Windows PowerShell, use the current working directory from the shell context:
py scripts/codexgo.py --cwd . --format json
Resume Work
Read these fields:
resolved_request: the best task to continue nowliteral_last_user_message: the exact last user message from the previous threadlast_conversation_content: the last meaningful conversation itemresolved_source: whether the request came from a user message, assistant suggestion, supplement, or fallbackneeds_more_context: whether the recovered text still looks ambiguousambiguity_hints: why the recovered text may still depend on earlier contextsupporting_context: nearby timeline entries, expanded upward when ambiguous references need older contextcontext_expanded_upward: whether the supporting context had to walk further upward for explanationdecision_basis_message: earlier user requirement that defines a selection or comparison basis
If the user wants execution, briefly state the recovered request and continue working immediately.
Trustgrade A
- passBody integrity
Whether the stored document is plausibly the kind of file the artifact declares, rather than something fetched by mistake.
- passType matchnot applicable to this artifact type
Whether the artifact is really the kind of thing its metadata claims it is.
- passFreshness
How long since the source repository was last pushed to.
- passPrompt injection
Scans the artifact's own text for instructions aimed at your agent rather than at you.
- passLicense
Whether the source repository declares an SPDX license permissive enough to redistribute.
How the grade is calculated
Each check contributes 0 points when it passes, 1 when it warns, and 2 when it fails. The total maps to a letter:
- Aevery check passed
- Bone warning
- Ctwo warnings
- Dprompt injection or body integrity failed, or three warnings
- Fone of those failed, and something else is wrong
These are automated hygiene checks, not a security audit, and not a dependency or vulnerability scan. A grade of A means nothing was flagged — not that the artifact is safe.
Versions
git-dab583fc37842026-07-31