agentleFS
Sign inStart free
CipherNovaaCipherNovaa/agent-skills/skills/myos/capture/SKILL.md

capture

Skill0 starsChanged 12 days agoMITView on GitHub

The natural-language front door — the user types a raw thought, you decide if it's a task or a note and file it through the right tool, confirming only when unsure. Use for any freetext capture that isn't already a /command.

57 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use weekly, ingest and 8 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

  • JaimeGBASkill

    ingest

    Digest an external source (article, paper, transcript, URL, book chapter) and integrate its content into the knowledge base

    Also in 8 projects that use this

  • AnthropicSkill

    brief

    Generate contextual briefings for legal work — daily summary, topic research, or incident response. Use when starting your day and need a scan of legal-relevant items across email, calendar, and contracts, when researching a specific legal question across internal sources, or when a developing situation (data breach, litigation threat, regulatory inquiry) needs rapid context.

    Also in 6 projects that use this

  • dubwoahSkill

    ask

    Answer questions about the codebase without generating or modifying any code. Use when the user wants to understand how something works, find where something is defined, or explore the project structure. Invoked with /ask.

    Also in 4 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

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 55 edited versions we read keep these.

  • 1

    Distinctive

    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 it came as /task / /note, skip classification — use that type.

    line 43
  2. Distinctive:

    Never lose it: If every branch fails, the Inbox is the floor — raw text always lands somewhere.

    line 62

“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

The natural-language capture layer. The user won't always reach for /task or /note — often they just type the thought as it leaves their head: "remind me to renew the domain monday", "idea: the janitor should report how long each check took". Your job is to catch it, decide what it is, and file it through the deterministic tool — so nothing is lost.

This is the fallback for when a /command wasn't used. If the user did use /task or /note, honour that type directly — don't re-classify.

The one rule that governs tone

  • Confident → just do it. File it, then a one-line receipt: "Done — task for Monday: renew the domain." No permission-seeking.
  • Unsure → one short question, then act. Ambiguous type or date → ask exactly one crisp question. If it's still unclear → drop it raw in the Inbox (below) and say so. Never guess a date; never lose the thought.

Classify (top-to-bottom, first match wins)

It looks like…Type → tool
an action to do — "remind me to…", "I need to…", "call/email/buy/fix…", esp. with a deadlinetask → myos_task.py add (see task skill)
a thought/idea/note-to-self with no action — "idea: …", "TIL …", "what if …"note → capture-to-inbox (verbatim, one line)
longer / needs a real home (a page, not a line)note → the /note command or capture-to-inbox; file-note sorts it later
a thing happening at a time you attend ("dentist Tuesday 2pm")event → your calendar tool if one is installed; otherwise a dated task
none clear after one questionInbox → capture-to-inbox (raw, flagged)

Signals: an imperative verb ("call", "buy", "email", "finish") or "remind me" → lean task. "idea/note/thought/TIL/what if" or pure musing → note.

Procedure

  1. If it came as /task / /note, skip classification — use that type.
  2. Otherwise classify with the table. Pull out the pieces:
    • task: terse imperative text, a --due <when> if there's a deadline, an --area if obvious. Pass the user's own date words ("monday", "the 21st") — the tool resolves them through myos_dates; never compute an ISO date yourself. No deadline → just add it undated.
    • note: the text, verbatim.
  3. Confidence check. Sure of both type and when? → file it, one-line receipt. Not sure? → one question. Still not sure? → capture-to-inbox raw, and say it's parked in the Inbox for later.

Examples

  • "remind me to renew the domain monday" → task with due date → myos_task.py add "Renew the domain" --due monday → "Noted — task for Monday."
  • "idea: janitor should time each check" → note → capture-to-inbox → "Stashed in the Inbox."
  • "squash update thing" → too vague → ask one question ("a task to do, or just a note?"); no clear answer → capture-to-inbox raw → "Wasn't sure, so I parked it in the Inbox."

Pitfalls

  • Never lose it. If every branch fails, the Inbox is the floor — raw text always lands somewhere.
  • Never guess a date. Ambiguous when → ask, or add the task undated.
  • Don't over-confirm. A clear task doesn't need a "shall I?" — the receipt IS the confirmation.
  • Don't re-classify an explicit /command. Trust the verb the user chose.
  • Let the tools own the files — you classify and hand off; you don't hand-write notes.

Verification

  • The item exists where its type belongs (Backlog line / Inbox line), created by its tool, and your receipt matches what the tool printed (its ✅/❌ line, relayed — not paraphrased).

Cadence

On demand — the default path for any freetext that isn't already a slash command.

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