record-knowledge
Skill4 starsChanged 5 months agoMITView on GitHub
Record tacit knowledge — quirks, pitfalls, dependencies, decisions, root causes — as tagged Markdown entries in `.claude/knowledge/entries/`. Use this skill whenever discoveries are made during work, when the user shares undocumented system behavior, or at plan completion to capture lessons learned. Also use when Claude Code makes a mistake pointed out by the user — record what happened, why it was wrong, and what to do next time.
In this repository3 other files it gives its agents.
- plan-taskSkill
- recall-knowledgeSkill
- review-knowledgeSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 19
New session start: Search .claude/knowledge/entries/ for active entries related to the current task before starting work
- Very specific: line 20
If an overview entry exists for the topic, read it first — load detail entries on demand to save context
- Distinctive: line 21
Progress update: Check if related entries need updating based on new discoveries
- Distinctive: line 22
Not needed when resuming a session (context is already preserved)
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Record Knowledge
Goal
Capture tacit knowledge discovered during work and make it available for future sessions.
When to Reference
- New session start: Search
.claude/knowledge/entries/for active entries related to the current task before starting work- If an
overviewentry exists for the topic, read it first — loaddetailentries on demand to save context
- If an
- Progress update: Check if related entries need updating based on new discoveries
- Not needed when resuming a session (context is already preserved)
Execution
Delegate the recording work to a subagent to minimize main context consumption.
- Analyze the user input or work context and prepare structured input fields
- Determine the knowledge base path (default:
.claude/knowledge/entries/) - Spawn a subagent with the following configuration:
Agent(
subagent_type: "general-purpose",
description: "record knowledge entry",
prompt: |
You are a knowledge recording agent. Read the procedure file and follow it precisely.
## Input
### what(事実)
{what}
### why(判断理由)
{why}
### context(背景情報)
{context}
### tags_hint(推奨タグ)
{tags_hint}
## Instructions
1. Read the procedure file at: {plugin_root}/skills/record-knowledge/procedure.md
2. Follow the procedure step by step
3. The knowledge base is at: {project_root}/.claude/knowledge/
4. Return: filename, # heading, ## headings, and linked entries
)Replace the placeholders:
{what}— 何が起きたか: the factual observation, discovery, or decision to record{why}— なぜ重要か: why this matters, the reasoning behind recording it (judgment made by the main agent){context}— 関連する背景情報: surrounding context (task being worked on, environment, related files, issues){tags_hint}— 推奨タグ: tags the main agent recommends (subagent validates against the tag registry){plugin_root}— the plugin's installation path (shown in the skill loading message as "Base directory for this skill"){project_root}— the project working directory
- Report the subagent's result to the user (filename, headings, linked entries only)
IMPORTANT: The procedure file path uses the plugin's base directory, NOT the project directory. Read the "Base directory for this skill" line from the skill loading message to determine the correct path.
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
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.

