Borrowing it
Nothing to install: this file belongs to IsmaelMartinez/delegate-local. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/IsmaelMartinez/delegate-local/main/CLAUDE.mdgit clone --depth 1 https://github.com/IsmaelMartinez/delegate-localWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/ismaelmartinez/delegate-local/claude-md)<a href="https://agentmods.dev/instructions/ismaelmartinez/delegate-local/claude-md"><img src="https://agentmods.dev/badge/instructions/ismaelmartinez/delegate-local/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/ismaelmartinez/delegate-local/claude-md"><img src="https://agentmods.dev/badge/instructions/ismaelmartinez/delegate-local/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.02947 | $0.02947 |
| Opus 5.5 | $0.01179 | $0.01179 |
| Sonnet 5.5 | $0.00589 | $0.00589 |
| Haiku 4.5 | $0.00295 | $0.00295 |
Grade B, and why
delegate-local CLAUDE.md scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
find scripts tests .claude/hooks -name '*.sh' -print0 | xargs -0 shellcheck -S error Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
There is no build step and no package manager. Run the whole suite (every file mocks `curl`, so a machine with a live model server gets the same counts as CI): How it starts
The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
What this repo is
This repo is a Claude Code skill, not an application that uses one. Editing SKILL.md changes how Claude itself behaves whenever the skill is loaded. Treat SKILL.md as production prompt content, not documentation: its frontmatter description is the prompt Claude reads to decide whether to invoke the skill, so an edit to it changes trigger accuracy in every conversation that has the skill installed. Keep the MUST/MUST NOT structure intact, and gate every description edit with the trigger eval below (recall ≥ 0.9 and negative-precision ≥ 0.9).
Since ADR 0035 (2026-10-18) the project is wound down to one shape: a large input Claude has not read (a long log, thread, document, directory or item list) is piped in chunks to a local model (MLX, Docker Model Runner or Ollama) over a shell pipe, and a short answer anchored to the input comes back for Claude to spot-check. Drafting text Claude will publish was measured net negative and left main with its hooks and calibration loop; docs/learnings.md keeps the lessons and commit a00b2e9 the code. There is no framework, router process or orchestration layer: the runtime path is scripts/pick-model.sh and scripts/delegate.sh. The discriminator is the local-brain insight: local models are strong summarisers and weak agents, so a task that needs multi-step reasoning, repo-wide context or tool-calling does not belong even if it looks textual. ROADMAP.md "Out of scope" enumerates the boundaries; do not add recipes or calibration machinery without a new ADR that answers ADR 0035's signposts.
Commands
There is no build step and no package manager. Run the whole suite (every file mocks curl, so a machine with a live model server gets the same counts as CI):
for f in tests/run-tests.sh tests/test-*.sh; do bash "$f"; done
Delegate, with or without a recipe (a recipe's tier comes from its frontmatter; --tier overrides):
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago Changed · -17 lines · -323 tokens per session 85a7ab5fd77b
- 20d ago Changed · +1 lines · +46 tokens per session 8edf2db15157
- 21d ago Changed · +2 lines · +111 tokens per session 55d455474a6d
- 22d ago Changed · +94 tokens per session d80c07d1a534
- 28d ago Changed · -50 lines · -7,690 tokens per session scan A → B 76dbfbc492fe
- 1mo ago Changed · +32 tokens per session 444add8bcc81
- 1mo ago Changed · -6 lines · +433 tokens per session 02837ad20931
- 1mo ago Changed · +627 tokens per session cae651b11404
- 1mo ago Changed · +7 lines · +827 tokens per session 07762df34112
- 1mo ago First seen · 158 lines · 8,790 tokens per session scan A e94763a592ee
delegate-local CLAUDE.md is an instructions file published in the GitHub repository IsmaelMartinez/delegate-local (8 stars, last pushed 9d ago), licensed MIT. It adds 2,947 tokens to every session, about $0.0118 per session on Opus 5.5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.