agentleFS
ovargasovargas/virtual-team/skills/token-efficient/SKILL.md

token-efficient

Skill14 starsChanged 8 months agoMITView on GitHub

Use when the user activates compressed output mode ("caveman mode", "terse", "compress") — drops conversational filler while preserving artifact quality. Deactivates on "normal mode" or "verbose". ADR-003 scopes this to conversation only.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When active, apply these rules to all conversational output (status updates, explanations, confirmations, progress reports, inline …

    line 20
  2. Distinctive:

    Rule of thumb: If it goes into a file or a git object, it is an artifact.

    line 40
  3. Distinctive:

    After the exception context passes, compression resumes without user action.

    line 51

≈ “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

Token-Efficient Communication Mode

Activation

On: User says "caveman mode", "terse mode", "terse", "compress", "be brief", or similar. Off: User says "normal mode", "verbose", "full prose", or similar.

Mode persists for the rest of the session or until deactivated. Acknowledge activation/deactivation in one short line:

  • On: Caveman mode on.
  • Off: Normal mode restored.

Compression Rules

When active, apply these rules to all conversational output (status updates, explanations, confirmations, progress reports, inline commentary):

  1. Drop articles (a, an, the) where meaning is preserved
  2. Drop filler phrases ("it is important to note", "in order to", "let me")
  3. Drop pleasantries ("Great question!", "Sure thing!", "Happy to help")
  4. Drop hedging ("I think", "it seems like", "probably", "might")
  5. Fragments OK — no need for complete sentences in status updates
  6. Technical terms exact — never abbreviate domain vocabulary, paths, identifiers, or numbers
  7. Tool calls unchanged — arguments to Edit, Write, Bash, etc. are never compressed

What Is NEVER Compressed (ADR-003)

This list is not configurable. These outputs use full prose regardless of mode:

  • Feature specs, implementation plans, reviews, bug reports
  • PR descriptions and commit messages
  • ADRs, CONTEXT.md, backlog entries
  • Any content written via Write or Edit tools (file content is an artifact)
  • Any output that persists beyond the current session

Rule of thumb: If it goes into a file or a git object, it is an artifact. Full prose.

Auto-Clarity Exception

Compression automatically suspends (full prose resumes) for:

  • Security warnings — vulnerabilities, credential exposure, permission issues
  • Irreversible action confirmations — destructive git operations, file deletions, database drops
  • Error explanations — when diagnosing a failure, precision matters more than brevity
  • Multi-step instructions — when compressed language risks the user misunderstanding the sequence

After the exception context passes, compression resumes without user action. Do not announce the switch.

Examples

Without mode (normal):

I've read the feature spec and identified the files that need to change. The main modification will be to the user service, where we need to add the validation logic. I'll also need to update the corresponding test file. Let me start with the test first, following TDD.

With mode (caveman):

Read spec. Changes: user service (add validation), test file. Starting with test (TDD).

Auto-Clarity Exception (even in caveman mode):

WARNING: This will force-push to main, overwriting 3 commits from other contributors. This cannot be undone. Proceed?

Integration

  • Triggered by: User activation phrase (mapped in skills/skill-awareness/SKILL.md)
  • Complements: skills/humanizer/SKILL.md (humanizer polishes artifacts; this compresses conversation)
  • Constraint: ADR-003 (docs/decisions/2026-04-29-token-mode-scope.md)

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.