skills/session-summary/SKILL.md
Use when you need to generate or view a session summary from work tracker logs.
npx skillsauth add seokan-jeong/team-shinchan team-shinchan:session-summaryInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Generate a quantitative session summary from work tracker data.
/team-shinchan:session-summary # Current session
/team-shinchan:session-summary --session session-xxx # Specific session
| Arg | Default | Description |
|-----|---------|-------------|
| --session {id} | current | Session ID to summarize (defaults to .shinchan-docs/.session-id) |
--session provided, use that ID.shinchan-docs/.session-id for the current session IDNo active session found. Use --session {id} to specify one.
Available sessions can be found via: /team-shinchan:work-log
.shinchan-docs/work-tracker.jsonl
No work tracker log found.
Events will be recorded automatically during Claude Code sessions.
{"ts", "type", "agent", "session", "data"}event.session matches the target session IDNo events found for session {id}.
From the filtered session events, calculate:
| Metric | How |
|--------|-----|
| Duration | First event ts to last event ts. Format as {start} - {end} ({N} minutes) |
| Agents invoked | Count unique non-null agent values from agent_start events. Build {agent: count} map. |
| Files changed | Count events where type === "file_change". Deduplicate by data.file path. |
| Delegations | Count events where type === "delegation". |
| Events total | Total count of all filtered events. |
Run git metrics to include development activity in the summary:
node ${CLAUDE_PLUGIN_ROOT}/src/analytics.js --git --format table
Append to summary output:
Git Activity:
Commits today: {N}
Lines: +{added} / -{deleted}
Files changed: {N}
Test ratio: {N}%
Hotspot files: {top 3 files}
Include in saved SESSION_SUMMARY.md as:
## Git Activity
| Metric | Value |
|--------|-------|
| Commits today | {N} |
| Lines changed | +{added} / -{deleted} |
| Files changed | {N} |
| Test ratio | {N}% |
| Hotspot files | {file1}, {file2}, {file3} |
Display the summary in console AND save to file.
Console output:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Session Summary
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Session: {session_id}
Duration: {start_time} - {end_time} ({N} minutes)
Agents invoked: {count} ({agent: count, ...})
Files changed: {count}
Delegations: {count}
Events total: {count}
Agent Activity:
| Agent | Invocations | Last Action |
|-------|-------------|-------------|
| {agent} | {count} | {last event summary} |
Key Events:
- {HH:MM}: {type} - {description}
...
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Determine output path:
.shinchan-docs/*/PROGRESS.md with status: active): .shinchan-docs/{doc_id}/SESSION_SUMMARY.md.shinchan-docs/SESSION_SUMMARY.mdWrite the summary using the same markdown template as hooks/session-wrap.md:
# Session Summary
- **Session**: {session_id}
- **Duration**: {start_time} - {end_time} ({N} minutes)
- **Agents invoked**: {count} ({agent: count, ...})
- **Files changed**: {count}
- **Delegations**: {count}
- **Events total**: {count}
## Agent Activity
| Agent | Invocations | Last Action |
|-------|-------------|-------------|
| {agent} | {count} | {last event summary} |
## Key Events
- {HH:MM}: {type} - {description}
Confirm:
Summary saved to {output_path}
session-wrap Stop hook but is user-triggered.shinchan-docs/work-tracker.jsonl.shinchan-docs/.session-idtesting
Default-on interview option-quality panel — N diverse generators produce structure-free options, a SelfCheckGPT majority-vote consensus filters hallucinations, a SteerConf cautious-confidence judge scores survivors, and a deterministic top-K is returned. Workflow tier; the single fierce-* skill that is ON by default.
development
Deterministic adversarial code review for high-stakes scope — independent per-dimension review, a non-skippable per-finding refutation, completeness + interaction critics, and a deterministic 3-lens rubric judge panel. Opt-in main-loop Workflow tier.
data-ai
Deterministic loop-until-done for high-stakes long-running tasks — a worker/verifier loop the script bounds by iteration cap, token budget, and stagnation, closed by an Action-Kamen gate. Opt-in main-loop Workflow tier.
testing
Deterministic adversarial debate for high-stakes or irreversible decisions — mandatory refutation plus a scored judge panel. Opt-in main-loop Workflow tier.