agentleFS
lethanhvietctt5lethanhvietctt5/claude-augment/skills/augment/SKILL.md

augment

Skill0 starsChanged 3 months agoMITView on GitHub

Rewrite a rough prompt into a stronger, structured one and hand it back for review without running it. Use when the user types /augment.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. qtzx06/yolodex★ 96same name
    2. shivae372/claude-bootstrap★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository3 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 3 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

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

    If that draft is empty, say Usage: /augment <your prompt> and stop.

    line 18
  2. Distinctive:

    Run the classifier, passing the draft on stdin via a heredoc so nothing needs escaping.

    line 22
  3. Very specific:

    If that command fails for any reason (no node, bad path), don't retry it — fall back to the prompt-leverage skill's tables and classify the …

    line 32
  4. Very specific:

    Follow the <augment-brief> the command returns.

    line 38
  5. Very specific:

    Do not wrap any of that in a > blockquote.

    line 46
  6. Distinctive:

    Keep the rewrite faithful.

    line 49
  7. Distinctive:

    Stop here: Do not start the work. Do not read files, search the codebase, or run commands for it.

    line 78

≈ “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.

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

Augment

Rewrite the draft and stop. The user gets the last look before any work starts. /augment-run is the variant that rewrites and then carries it out.

The user's draft prompt:

$ARGUMENTS

If that draft is empty, say Usage: /augment <your prompt> and stop.

Step 1 — get the brief

Run the classifier, passing the draft on stdin via a heredoc so nothing needs escaping. Run the command unquoted, exactly as written below, so it matches the pre-approved permission rule:

node ${CLAUDE_PLUGIN_ROOT}/bin/augment.mjs brief <<'AUGMENT_DRAFT_EOF'
<the draft, verbatim>
AUGMENT_DRAFT_EOF

If that command fails for any reason (no node, bad path), don't retry it — fall back to the prompt-leverage skill's tables and classify the draft yourself. Say in one line that you fell back, then continue.

Step 2 — rewrite

Follow the <augment-brief> the command returns. It resolves the intent, the rewrite mode, and the effort level, and carries the rules for the rewrite.

Print exactly two things, in this order, and nothing else in this step:

  1. One header line, on its own: **augment** · <intent> · <mode> · <effort>
  2. The enhanced prompt, inside a single fenced code block.

Do not wrap any of that in a > blockquote. A fenced code block nested inside a blockquote renders broken — stray > markers end up inside the prompt.

Keep the rewrite faithful. You are restructuring the user's request, not answering it and not expanding its scope.

Step 3 — put it on the clipboard

Claude Code has no API for writing its own input box, so the rewrite is handed over via the system clipboard instead. Run this once, unquoted, with the rewrite you just printed on stdin:

node ${CLAUDE_PLUGIN_ROOT}/bin/augment.mjs copy <<'AUGMENT_REWRITE_EOF'
<the enhanced prompt, verbatim>
AUGMENT_REWRITE_EOF

Send only the enhanced prompt — no header line, no code fence, nothing the user would have to delete after pasting.

The command prints one line and always exits 0. Close on whichever it was:

  • copied to clipboard via … → Paste it into the prompt box to run it, or /augment-run <draft> to rewrite and run in one step.
  • clipboard unavailable: … → say the clipboard could not be reached, repeat the reason verbatim, and tell the user to copy the block above by hand.
  • clipboard hand-off is disabled … → say nothing about it; just point at the block above.

Step 4 — stop

Stop here. Do not start the work. Do not read files, search the codebase, or run commands for it. Do not ask whether to proceed — asking is not stopping. The whole point of this command is that nothing happens until the user sends the prompt themselves.

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.