agentleFS
Sign inStart free
viasrijanviasrijan/crazy-skills/skills/productivity/token-efficiency/SKILL.md

token-efficiency

Skill0 starsChanged 3 months agoMITView on GitHub

Use when you want to minimize token usage — efficient file reads, command execution, output handling, and model selection for cost-effective sessions.

  • Installs packages

12 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use tdd.

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.
  • GoogleSkill

    tdd

    Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 12 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Default for all sessions — apply without being asked unless user says "verbose"

    line 12
  2. Very specific:

    Prefer Grep/Glob + Bash filters over Read entire files.

    line 23
  3. Very specific:

    Use --quiet/--silent, head/tail sampling for large output.

    line 25

“Common” is measured: kept by at least 6 in 10 of the 12 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

Token Efficiency

Token optimization for cost-effective sessions. From Delphine-L/claude_global token-efficiency (90-95% weekly saving) + Headroom integration.

When to use

  • Default for all sessions — apply without being asked unless user says "verbose"
  • Reading log files, large outputs, codebase navigation, debugging, system status

Workflow

Model routing (50% saving vs all-Opus):

  • Opus: 10-15 min learning a new codebase (architecture, patterns)
  • Sonnet: all implementation, debugging, routine work (default)
  • Return to Opus only for deep architectural understanding or when Sonnet fails

Reading & execution (saves 70-80% vs full reads):

  1. Prefer Grep/Glob + Bash filters over Read entire files. Example: Grep "class Foo", head -100, jq '.metadata', head -20 for CSV + wc -l.
  2. For exploratory searches, dispatch a task subagent — main context gets a paragraph back, not 40K tokens.
  3. Use --quiet/--silent, head/tail sampling for large output. Skills themselves are progressive disclosure: ~155 tokens per skill at start, body loads only when triggered.
ApproachTokens/Week
Wasteful (Read everything)500K
Moderate (filtered reads)200K
Efficient (bash + filters + subagents)30-50K

Headroom (optional plugin): npm install @ngotrnghia1811/opencode-headroom and add to opencode.json plugin — 60-95% on large tool outputs with local cache (CCR).

Verification

  • Default model is Sonnet; Opus only for learning/deep analysis
  • Large files read via filters/subagents, not full Read
  • Large command outputs sampled with head/tail/jq
  • Skills body not assumed in-context — only frontmatter at start

References

  • Source: Delphine-L/claude_global token-efficiency
  • See also: productivity/prompt-enhancement for input optimization, productivity/context-compression for long sessions

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