with one click
session-log
// Inspect prove_it session logs—review history, reviewer verdicts, and session state. Use when asked about session logs, reviewer results, or "what did the last hook say".
// Inspect prove_it session logs—review history, reviewer verdicts, and session state. Use when asked about session logs, reviewer results, or "what did the last hook say".
Release a new version of prove_it. Use when asked to cut a release, publish a new version, or "ship" it.
Create a temporary real project and prove a prove_it feature works (or doesn't) end-to-end. Builds a disposable git repo, writes a focused config, runs real dispatches through the installed or local prove_it, and produces a human-readable session transcript. Use when you need to prove a feature, reproduce a bug, or validate a fix against a real project — not just unit tests.
Test prove_it from source against an example project or temp directory. Use when you want to verify hook behavior end-to-end without a Homebrew release.
| name | session-log |
| description | Inspect prove_it session logs—review history, reviewer verdicts, and session state. Use when asked about session logs, reviewer results, or "what did the last hook say". |
Session data lives in ~/.claude/prove_it/sessions/. Two file types per session:
<session_id>.json—session state (project dir, git HEAD, started_at)<session_id>.jsonl—reviewer log (one JSON object per line: at, reviewer, status, reason, projectDir, sessionId)_project_<hash>.jsonl—project-level reviewer log (when session_id was null)The current Claude Code session ID is passed in hook input. To find it for the active conversation, check the most recently modified .jsonl file:
ls -t ~/.claude/prove_it/sessions/*.jsonl | head -5
.jsonl file (excluding _project_* and test-session* files).at timestamp)Read <session_id>.jsonl and <session_id>.json and display both the session state and reviewer entries.
--all or all: list all sessionsList all session .json files sorted by modification time, showing session ID, project, and started_at.
--stats or stats: aggregate reviewer statsCount PASS/FAIL/SKIP/BOOM across all .jsonl files and show totals.