continue-codex-work
An agent skill by daymade, from daymade/claude-code-skills. Tags: analysis, local-workflow, resilience, triage, workflow.
What it does
Continues interrupted OpenAI Codex work only after read-codex-history verifies the selected rollout identity and complete fork/compaction lineage. Use when the user provides a Codex Session ID, asks to pick up a prior Codex run, says Codex was interrupted, fused, compacted, or stuck, or wants the current Agent to take over without `codex resume`. Restores the original business outcome, unfulfilled requests, user corrections, proven prior assets, current workspace truth, and the next action that directly advances the goal. Do not use when Codex itself natively resumed this same conversation and its prior turns or compaction state are already present in the current context.
Install
With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:
npx skills add daymade/claude-code-skills --skill continue-codex-workOr copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):
git clone --depth 1 https://github.com/daymade/claude-code-skills
cp -r claude-code-skills/daymade-claude-code/continue-codex-work ~/.claude/skills/continue-codex-workSafety box score
Grade B, low risk: 14 of 100 risk points, rated 2026-10-20. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.
This is a meta-skill for resuming an earlier Codex session by reading session history, reconciling current workspace state, and then continuing the unfinished task. Its main risks are bounded read access to project/session data and the autonomy to carry forward whatever action the prior task required; it does not ship executable code or privilege changes.
| Category | Level | Why |
|---|---|---|
| Code & shell execution | 0 (none) | No shell commands, code execution, or bundled scripts are present; it only instructs the agent to read history and project files. |
| Network access | 1 (low) | It invokes `daymade-claude-code:read-codex-history` for the exact Session ID, which implies a bounded read from a history service, but not general data exfiltration. |
| Filesystem reach | 1 (low) | It asks to read `AGENTS.md`/`CLAUDE.md`, check cwd/branch/working tree, and inspect relevant files, but does not write or delete files itself. |
| Secrets & credentials | 0 (none) | The skill asks for a session ID and history briefing, but it never requests passwords, tokens, key files, or `.env` contents. |
| Privilege & persistence | 0 (none) | No sudo, service registration, shell startup edits, or global installs are requested. |
| Instruction hijack surface | 0 (none) | The instructions are a normal workflow contract; they do not tell the agent to ignore higher-priority rules, hide actions, or follow untrusted fetched content. |
| Irreversible actions | 2 (moderate) | It tells the agent to take over an earlier rollout and to "Perform the next direct action and close its feedback loop," so it can carry forward whatever business action the inherited task requires. |
| Transparency | 0 (none) | All shipped content is readable markdown; no binaries, encoded blobs, or downloaded installers are present. |
Before installing
- Confirm `daymade-claude-code:read-codex-history` is read-only and cannot mutate goals or fetch secrets.
- Read the target project's `AGENTS.md` / `CLAUDE.md` first; this skill will follow those rules.
- Test it in a scratch session ID and verify it fails closed on mismatched or ambiguous history.
Rated 2026-10-20 by an automated review of 2 files from the skill's repository.
Source
- Repository
- daymade/claude-code-skills (all skills from this repository)
- Path
- daymade-claude-code/continue-codex-work/SKILL.md
- Branch
- main
- Updated
- 2026-09-20
Related skills
- continue-claude-code-work — Continues interrupted Claude Code work only after local history has been read and reconciled.
- triage-issue-reports — Triage Slack issue reports with one thread-only verdict, evidence review, cause-aware routing, tracker dedupe, and fail-closed ticket creation.
- local-conversation-history — Entry point for local AI conversation history across providers.
- br — Official skill for beads_rust (`br`), a local-first, dependency-aware issue tracker for AI agents.
- resilience-hub-failure-mode-assessment — Runs and interprets AWS Resilience Hub v2 failure mode assessments.
- automate-me — Use for "automate me", "create/update/refresh my -mode skill", "turn/capture my preferences or working style into a skill".