agentleFS
aiflaredevaiflaredev/aiflare/skills/summarize/SKILL.md

summarize

Skill0 starsChanged 6 months agoMITView on GitHub

Generate a summary report for a Claude Code session, capturing what was done, why, and what alternatives were considered.

93 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use openclaw-performance, plugin-sdk-impact and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • clawdbotSkill

    openclaw-performance

    Performance work in OpenClaw: a slow or heavy flow (latency, CPU, memory, SQL, RPC, renders), writing or reviewing a perf PR, or adding a cache. Count the work, delete waste first, cache last.

    Also in 7 projects that use this

  • clawdbotSkill

    plugin-sdk-impact

    Count Plugin SDK consumers across live ClawHub releases and bundled OpenClaw plugins before deprecation, removal, or behavior changes; supports public release-date filters.

    Also in 7 projects that use this

  • AtomicBot-aiSkill

    taskflow-inbox-triage

    Also in 6 projects that use this

  • hkyutongSkill

    node-connect

    Diagnose OpenClaw node connection and pairing failures for Android, iOS, and macOS companion apps. Use when QR/setup code/manual connect fails, local Wi-Fi works but VPS/tailnet does not, or errors mention pairing required, unauthorized, bootstrap token invalid or expired, gateway.bind, gateway.remote.url, Tailscale, or plugins.entries.device-pair.config.publicUrl.

    Also in 18 projects that use this

  • molt-botSkill

    discord-e2e

    Test Discord with Convex-leased QA bots: check credentials without a Gateway, exercise owned messages, or prove OpenClaw replies, files, threads, reactions, streaming, and typing. Distinguish API and Gateway evidence from manual user interactions.

    Also in 10 projects that use this

  • molt-botSkill

    openclaw-repair-sweep

    Run scoped OpenClaw issue/PR repair campaigns: coordinate workers, prove root causes, and land or close verified work under the requested authority.

    Also in 10 projects that use this

  • +2 more from molt-bot

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 7 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 81 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before generating the report, run git log --oneline -3 to detect the language used in recent commit messages.

    line 10
  2. Very specific:

    Call the get_session_summary MCP tool to retrieve session data.

    line 14
  3. Very specific:

    If a session ID is provided as an argument ($ARGUMENTS is not empty), pass it as sessionId.

    line 15
  4. Very specific:

    If no argument is given, omit sessionId (the current session will be used automatically).

    line 16
  5. Very specific:

    Important: Always insert the HTML comment marker <!-- CONTINUATION_DIRECTIVE_START --> immediately before the Continuation Directive …

    line 28
  6. Distinctive:

    If the session has incomplete work, use this template:

    line 30
  7. Distinctive:

    If all planned work is complete, replace the entire template with:

    line 55
  8. Distinctive:

    Be specific: Write "Implement DELETE /api/v1/feedbacks/{id} endpoint" instead of "Implement API" — the agent should be able to act …

    line 66
  9. Distinctive:

    Include rationale: Every decision and task item must explain "why".

    line 67
  10. Distinctive:

    Use present tense: Write "X must be done" not "X was planned" — frame as current instructions, not past plans

    line 68
  11. Distinctive:

    Empty fields: Write "None" but keep the field to maintain structural consistency

    line 69
  12. Very specific:

    Call the save_session_report MCP tool to save the report to the server.

    line 78

“Common” is measured: kept by at least 6 in 10 of the 81 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Generate a session summary report.

Writing Language

Before generating the report, run git log --oneline -3 to detect the language used in recent commit messages. Write the entire report in the same language. For example, if commit messages are in Korean, write the report in Korean. If in English, write in English.

Instructions

  1. Call the get_session_summary MCP tool to retrieve session data.
    • If a session ID is provided as an argument ($ARGUMENTS is not empty), pass it as sessionId.
    • If no argument is given, omit sessionId (the current session will be used automatically).
  2. Based on the returned data, generate a summary report in Korean with the following sections:
    • Title: A one-sentence summary of the session's key work
    • Overview: 2-3 sentences describing the overall workflow of the session
    • Key Changes: For each capture entry, summarize what was done and why
    • Rejected Alternatives: Include alternatives if present
    • Changed Files: List all changed files
    • Continuation Directive: A structured directive for the next session's agent to continue the work. Generate using the template and guidelines below.

Continuation Directive Template

Important: Always insert the HTML comment marker <!-- CONTINUATION_DIRECTIVE_START --> immediately before the Continuation Directive heading. This marker is used by the frontend to extract and display the directive separately.

If the session has incomplete work, use this template:

<!-- CONTINUATION_DIRECTIVE_START -->
## Continuation Directive

> **Goal**: The end goal this session was working toward (one sentence)
> **Current State**: What has been completed so far (one sentence)
> **Completion**: Approximate percentage (e.g., ~60%)

### Remaining Tasks
- [ ] Specific incomplete item

### Recommended Order of Work
1. What to do first and why

### Design Decisions to Preserve
| Decision | Rationale |
|----------|-----------|
| What was decided | Why it was decided this way |

### Caveats
- Constraints or pitfalls the agent must be aware of

If all planned work is complete, replace the entire template with:

<!-- CONTINUATION_DIRECTIVE_START -->
## Continuation Directive

All planned work for this session has been completed.

Directive Quality Guidelines

  • Be specific: Write "Implement DELETE /api/v1/feedbacks/{id} endpoint" instead of "Implement API" — the agent should be able to act immediately
  • Include rationale: Every decision and task item must explain "why". Without rationale, the agent may arbitrarily change direction
  • Use present tense: Write "X must be done" not "X was planned" — frame as current instructions, not past plans
  • Empty fields: Write "None" but keep the field to maintain structural consistency
  • Data sources:
    • Goal / Current State: synthesize from entries' intent fields
    • Remaining Tasks: infer planned items not yet completed from entries
    • Design Decisions: extract "chosen vs rejected" from entries' alternatives fields
    • Caveats: extract constraints from entries' intent/alternatives fields
  1. Display the generated report to the user.
  2. Call the save_session_report MCP tool to save the report to the server.
    • title: The title generated above
    • content: The full report content (Markdown)
    • sessionId: The same value used in step 1
  3. Display a save confirmation message.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.