AgentSkillsCN

compound-learnings

将近期解决的问题记录为可复用的解决方案，存入docs/solutions/目录。在完成一个问题后，记录问题本身、根本原因、解决方案以及预防策略，以便未来的代理与开发者从中汲取经验教训。

SKILL.md
--- frontmatter
name: compound-learnings
description: >
  Document a recently solved problem as a reusable solution in docs/solutions/.
  Use after completing an issue to capture the problem, root cause, solution, and
  prevention strategy so future agents and developers can learn from it.
argument-hint: "[path to completed issue or description of solved problem]"

Compound Learnings

Pipeline Role

Step 5 of the connected pipeline: Capture → Plan → Work → Review → Compound.

This skill closes the knowledge loop. When a problem is solved, it documents the solution so future work can reference it. This is the mechanism that makes the system smarter over time.

When to Use

Activate when:

  • •An issue has been resolved and the solution should be documented
  • •A tricky bug was fixed and the root cause should be remembered
  • •A pattern or approach was discovered that would help with similar future problems
  • •The user explicitly wants to document a learning

Steps

1. Gather the Learning

If an issue file is provided, read it for context. Otherwise, ask the user:

  • •What was the problem? (symptom)
  • •What caused it? (root cause)
  • •How was it fixed? (solution)
  • •How can it be prevented? (prevention)

2. Categorize

Choose the most appropriate category for docs/solutions/:

  • •performance-issues/ — slow queries, memory leaks, scaling problems
  • •security-issues/ — vulnerabilities, auth bugs, data exposure
  • •build-errors/ — dependency issues, compilation failures, CI problems
  • •configuration-fixes/ — environment config, deployment issues, integration problems

If none fit, create a new category directory.

3. Create Solution File

Path: docs/solutions/<category>/<descriptive-slug>.md

yaml
---
title: "[Descriptive title of the problem and solution]"
date: YYYY-MM-DD
category: [category-name]
tags: [relevant, technology, tags]
module: [affected module or area]
symptom: "[What the developer observed]"
root_cause: "[Why it happened]"
severity: low|medium|high|critical
---

## Problem
[Detailed description of what went wrong]

## Root Cause
[Technical explanation of why it happened]

## Solution
[What was done to fix it, with code snippets where helpful]

## Prevention
[How to avoid this in the future — tests, linting rules, conventions]

4. Update Agent Context

If the learning reveals a pattern that future agents should know about, append a brief note to .github/agent-context.md:

markdown
### [Category]: [Brief finding]
[One-sentence summary with reference to docs/solutions/<file>]

5. Update Plan File

If working from a plan file, set status: done and append a final activity entry:

markdown
### YYYY-MM-DD HH:MM — Issue completed
- Learning documented: `docs/solutions/<category>/<file>.md`
- Agent context updated: [Yes/No]
- **Status:** Done

6. Print Summary

Confirm: "Learning documented at docs/solutions/<path>. Future agents will reference this when encountering similar problems."

Guardrails

  • •Keep solution files focused — one problem, one solution per file.
  • •Include enough context that someone unfamiliar with the codebase can understand the learning.
  • •Use specific code examples, not vague descriptions.
  • •Tag accurately — tags are how future agents find relevant learnings.