agentleFS
lingengyuanlingengyuan/my-skills/.claude/skills/insight-collector/SKILL.md

insight-collector

Skill9 starsChanged 3 days agoMITView on GitHub

Analyze URLs, source code, images, technical documents, or notes, and archive reusable insights, code snippets, and architecture patterns into a knowledge base. Use when collecting knowledge, archiving technical articles, saving reusable code, or recording insights (收录, 记录一下).

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 9 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If the target directory is not specified, inspect the workspace for existing catalogs (e.g.

    line 15
  2. Distinctive:

    Analyze source material: Read the source thoroughly to understand core arguments, architecture decisions, and boundary conditions.

    line 20
  3. Distinctive:

    Keep source facts and analytical inferences strictly distinguished.

    line 31
  4. Distinctive:

    Update index catalog: If the knowledge base maintains an index table in README.md, add a row referencing the new asset.

    line 34
  5. Distinctive:

    Verify and report: Verify that all files exist and links resolve.

    line 35
  6. Distinctive:

    Language consistency: Write analytical notes in clear Chinese unless the target project specifies English.

    line 39
  7. Distinctive:

    Fact preservation: Never fabricate claims, fake benchmark results, or code that was not present in or derived from the source.

    line 40
  8. Distinctive:

    No pseudo-profundity: Do not force artificial "mental models", pretentious philosophizing, or fabricated knowledge-base links onto …

    line 41
  9. Distinctive:

    Directory isolation: Never write outside the target knowledge base directory.

    line 43
  10. Distinctive:

    Before completing the task, verify:

    line 47

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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.

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

Insight Collector Skill

Extract actionable engineering insights, reusable code snippets, and system design patterns from raw material and archive them into structured knowledge assets.

Inputs

  • Source Material: URL, code repository, image, screenshot, PDF, or text notes.
  • Target Knowledge Base: Target directory for archiving.

If the target directory is not specified, inspect the workspace for existing catalogs (e.g. README.md, INDEX.md, READING_LIST.md). If multiple plausible destinations exist, ask the user to select the location before writing.

Workflow

  1. Inspect existing catalog: Inspect the target directory to understand existing taxonomy and prevent duplicate notes.
  2. Analyze source material: Read the source thoroughly to understand core arguments, architecture decisions, and boundary conditions.
  3. Extract high-signal items:
    • Reusable code: Self-contained functions, scripts, and configuration blocks with explicit dependency requirements.
    • System design patterns: Concrete architecture choices, component boundaries, and data flows.
    • Trade-offs and failure modes: Real-world constraints, performance bottlenecks, and edge cases.
    • Actionable implications: Concrete follow-up ideas or spike experiments.
  4. Structure the document:
    • Select an appropriate document template from references/doc-templates.md:
      • In-depth technical analysis (analysis/<topic>.md)
      • Project inspiration note (ideas/<topic>.md)
      • Standalone snippet (python/, javascript/, shell/)
    • Keep source facts and analytical inferences strictly distinguished.
    • Include original source URL or file path and capture date.
  5. Categorize and save: Save the file in the matching category folder. Follow references/project-conventions.md when the target project adheres to those conventions.
  6. Update index catalog: If the knowledge base maintains an index table in README.md, add a row referencing the new asset.
  7. Verify and report: Verify that all files exist and links resolve. Present the key takeaways and saved paths concisely.

Guidelines & Boundaries

  • Language consistency: Write analytical notes in clear Chinese unless the target project specifies English. Keep code and technical identifiers in their original programming language.
  • Fact preservation: Never fabricate claims, fake benchmark results, or code that was not present in or derived from the source.
  • No pseudo-profundity: Do not force artificial "mental models", pretentious philosophizing, or fabricated knowledge-base links onto straightforward technical material. Keep notes pragmatic.
  • Minimal experiments: Proposed follow-up spikes must state source motivation, expected value, and minimal verification criteria.
  • Directory isolation: Never write outside the target knowledge base directory.

Verification Checklist

Before completing the task, verify:

  1. The generated markdown file exists and contains valid frontmatter or headers.
  2. Source link and capture date are accurately recorded.
  3. Code blocks specify language identifiers and required dependencies.
  4. Existing catalog or index file is updated if applicable.
  5. Output summary reports file paths and key findings directly.

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.