The interface is localized. Skill descriptions and documents are machine-translated by AI into your language; code, commands, paths and technical terms stay in the original, and the author's document governs.
Explore skills
Development · SKILL.md

run-history

run-history

Inspect CCAM dashboard-run history and relate live run handles to persisted Claude Code or Codex sessions. Use when finding a prior launched task, checking whether a run is still attached, reviewing start/end status, or deciding whether to resume, view, or relaunch work.

Independent GitHub snapshot · Automated classification · Original documentation governs

Skill documentation

GitHub

name: run-history description: > Inspect CCAM dashboard-run history and relate live run handles to persisted Claude Code or Codex sessions. Use when finding a prior launched task, checking whether a run is still attached, reviewing start/end status, or deciding whether to resume, view, or relaunch work.

Run History

  1. Run ccam run history --limit 100.
  2. Filter by provider, working directory, model, status, and session ID.
  3. For a live handle, run ccam run get <id> --envelopes.
  4. For the persisted agent session, use ccam session <session-id> and the session transcript tools.
  5. Distinguish:
    • process still live
    • process completed with usable output
    • process failed
    • history row exists but the in-memory handle was reaped
  6. Recommend resume only when the native session ID is present and the selected provider supports that session.

Return the exact run ID, provider, session ID, working directory, model, timestamps, lifecycle state, and the next safe action.

Saved skills