agentleFS
Sign inStart free
playingpossummplayingpossumm/loose-ends/.claude/skills/capture/SKILL.md

capture

Skill0 starsChanged 33 days agoMITView on GitHub

File something into the vault inbox, whether a link, a PDF, an image, a thought, or the current Claude Code session. Use when the user says "capture this", "file this", "save this to my brain", or wants to keep something from the conversation. Does not compile; that is /ingest.

57 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use weekly, weekly-review and 5 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.
  • gaiaaiagentSkill

    weekly

    Generate a weekly digest rolling up dailies into higher-level pattern analysis

    Also in 4 projects that use this

  • QuixiAISkill

    weekly-review

    Review the past week across goals, tasks, calendar, journal, and commitments, then propose the next week

    Also in 5 projects that use this

  • termdockSkill

    recall

    Retrieves cross-session memories about past decisions, lessons, and patterns. Use when user asks about prior work, past decisions, or mentions a feature. Triggers include check memory, what did we decide, how did we solve.

    Also in 6 projects that use this

  • AlexPEClubSkill

    refine

    Always use when the user wants to discuss an exsiting feature or specification. Open an existing feature spec to improve, extend, or fundamentally challenge it. Pass the feature ID as argument (e.g. /refine PROJ-2).

    Also in 3 projects that use this

  • LangChainSkill

    remember

    Review the current conversation and capture valuable knowledge — best practices, coding conventions, architecture decisions, workflows, and user feedback — into persistent memory (AGENTS.md) or reusable skills. Use when the user says: (1) remember this, (2) save what we learned, (3) update memory, (4) capture learnings.

    Also in 3 projects that use this

  • VercelSkill

    setup

    Set up Vercel CLI and project configuration. Use when the user says "set up Vercel", "configure Vercel", "link to Vercel", or "vercel init".

    Also in 8 projects that use this

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 55 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

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

    Put something into raw/inbox/ so /ingest can compile it later.

    line 8
  2. Very specific:

    Write raw/inbox/<id>.md with this frontmatter, then the content verbatim:

    line 31
  3. Distinctive:

    Preserve wording: Never summarise or clean up at capture time.

    line 52
  4. Very specific:

    Do not compile: Do not write to wiki/, loops/, or index.md.

    line 55
  5. Distinctive:

    Never ask more than one clarifying question, and only if you genuinely cannot tell what to file.

    line 60

“Common” is measured: kept by at least 6 in 10 of the 55 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

capture

Put something into raw/inbox/ so /ingest can compile it later. Capture is deliberately dumb and fast, in that it files without thinking. Friction here kills the whole system, so never interrogate the user before filing.

This is one of several doors into raw/inbox/, alongside the Telegram script, the Obsidian Web Clipper, brain_capture over MCP, and plain drag-and-drop. They all land in the same folder and produce the same shape of file, so /ingest never needs to care which was used.

What to file

The argument, or if there is none, infer from context:

InputAction
URLfetch it, save as markdown with the original URL in frontmatter
Local file pathcopy into raw/inbox/, keep the original name
Image / screenshotcopy in, and write a text transcription alongside it. The transcription is what gets compiled, and the image stays as provenance
Free text / a thoughtwrite it verbatim as a note. Preserve the user's wording.
"this session" / "this conversation"write the substance of the conversation so far as a transcript

How

  1. Generate an id: YYYY-MM-DD-<short-slug>.
  2. Write raw/inbox/<id>.md with this frontmatter, then the content verbatim:
---
id: <id>
captured: YYYY-MM-DD
kind: article | pdf | image | note | transcript
origin: <url, file path, or "conversation">
title: <short human title>
status: uncompiled
---
  1. For images: save the image file beside it as <id>.<ext>, add image: <id>.<ext> to the frontmatter, and put your transcription in the body. Note in the body that the text is a transcription, so /ingest cites the image rather than treating the text as original.
  2. Confirm in one line: what was filed, its id, and that it is uncompiled.

Rules

  • Preserve wording. Never summarise or clean up at capture time. Compilation is where interpretation happens; capture is a photocopier.
  • Never edit anything already in raw/. It is immutable.
  • Do not compile. Do not write to wiki/, loops/, or index.md. That is /ingest.
  • If it already exists (same content hash or same origin URL), say so and do not duplicate, but note the re-encounter in the existing file's body with today's date. A repeat encounter is a signal that the material matters.
  • Refuse passwords, API keys, credentials, and health records. Say you refused and why.
  • Never ask more than one clarifying question, and only if you genuinely cannot tell what to file.

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.

Check your setup, free