agentleFS
preetamnathpreetamnath/agent-skills/skills/compress-file/SKILL.md

compress-file

Skill4 starsChanged 7 months agoMITView on GitHub

Compress one instruction file to the leanest STRUCTURE that still delivers its purpose — dissolve sections that restate others, fold unique survivors into the section that governs them. TRIGGER when: user says 'why is this so long', 'strip the fat', 'de-duplicate this skill/doc', 'these sections overlap'.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    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:

    Write Nothing to compress — already lean.

    line 38

≈ “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

Compress File

Cut structure, not meaning — the file must still deliver its purpose, in fewer lines.

Steps

Step 1 — Pin the purpose + its invariants

State in one line what the file is for. List every distinct instruction it must still deliver — the checklist you prove against in Step 4.

Step 2 — Find and remove structural fat

Map which sections restate the purpose, the description, or each other. Judge each overlap:

  • CUT — content another section or the description already carries.
  • FOLD — unique content in the wrong home → move it into the section that governs it, then delete the emptied section.

Score each finding 0.00–1.00 — how sure you are the CUT loses nothing, or the FOLD's new home is right.

invoke the tighten-instruction skill via the Skill tool; tighten each survivor in place. Leave the file's output-contract or skeleton near-untouched — it's the product.

Step 3 — Score-gate and apply

Show the plan, then Edit every finding scoring ≥ 0.80; list the rest as held, un-applied. The edits land in a reviewable diff, revert in one commit, and Step 4 re-reads them cold — so the floor applies without walking each.

**Compression plan — <file>**
Purpose (must survive): <one line>
- [0.92] CUT:  <section/line> — already in <where>
- [0.60] FOLD: <content> → <target section>  · held (< 0.80)

Write Nothing to compress — already lean. when there's nothing to change.

Step 4 — Prove and quantify

Re-read each changed section cold, plan out of view. Confirm every Step 1 invariant survived — in place and still reading right — or was an approved CUT; a FOLD can strand a referent or half a rule, so revert or fix any that broke. Then report net lines and words removed, and the percentage.

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.