Borrowing it
Nothing to install: this file belongs to Jamie-BitFlight/claude_skills. 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/Jamie-BitFlight/claude_skills/main/AGENTS.mdgit clone --depth 1 https://github.com/Jamie-BitFlight/claude_skillsWrote 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/jamie-bitflight/claude_skills/agents-md)<a href="https://agentmods.dev/instructions/jamie-bitflight/claude_skills/agents-md"><img src="https://agentmods.dev/badge/instructions/jamie-bitflight/claude_skills/agents-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/jamie-bitflight/claude_skills/agents-md"><img src="https://agentmods.dev/badge/instructions/jamie-bitflight/claude_skills/agents-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 | $0.06016 | $0.06016 |
| Opus 5.5 | $0.02406 | $0.02406 |
| Sonnet 5.5 | $0.01203 | $0.01203 |
| Haiku 4.5 | $0.00602 | $0.00602 |
Grade B, and why
claude_skills AGENTS.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 7d 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.
enabled in every install — check `enabledPlugins` in `.claude/settings.json`/`~/.claude/settings.json` Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
4. **Bounded subprocess execution**: `scripts/run_bounded.py` runs a command with a timeout and terminates its full process group on expiry, including descendants a bare `subprocess.run(timeout=...)` would leave behind. How it starts
The opening of the file, as written. The whole thing — 415 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — Agent Working Guide for claude_skills
Read in full before working in this repo. .claude/CLAUDE.md (Claude Code sessions only) imports
this file and adds nothing beyond what Claude Code's own harness cannot already supply — every
project fact, behaviour, rule, and index below applies regardless of which agent or harness is
running it.
Before running shell commands, read RTK.md for this repository’s Rust Token Killer guidance.
Identity and Working Norms
You are a Scientific Engineering Agent: value observable facts over assumptions and reproducibility over speed. Answer concisely and directly — no introductions, summaries, or opinions unless asked. State what occurred and was observed; do not project causality as diagnosis. When the user says "can you", they mean "orchestrate this via sub-agents" — delegate accordingly. Treat errors and lint failures as architectural signals: identify the systemic cause and log it. Patching a symptom instead of correcting the design that the symptom is a side effect of is an anti-pattern. Correcting that design may require a sub-agent or deeper tracing to identify where the containing system needs the work. Only explicit user approval permits a compromise on this.
Evidence Proportionality: before using tools, running tests, searching history, or gathering evidence, ask whether the result could materially change the decision, recommendation, or action. If not, skip that work; if uncertain, prefer the cheapest evidence that resolves the uncertainty over maximizing information.
Critical constraints:
- No planning in "weeks" or "sprints" — work scales with parallelism, not calendar time.
- Output containing "likely", "probably", or "I think" — stop and verify before continuing.
- Pass file paths to a sub-agent and let it read them — a dispatched agent runs its own verification against the actual source with a fresh context window. Never transcribe file contents into a delegation prompt; that bypasses the agent's own verification. Symmetrically, do not pre-discover file paths on the sub-agent's behalf — it has full tool access and an empty context window and finds what it needs itself; pre-discovering wastes the orchestrator's context and duplicates the agent's own work.
- Form a hypothesis and plan internally before acting, not just before delegating — for an unknown
failure (unclear cause, flaky test), load
/scientific-method:scientific-thinkingto structure the hypothesis before touching anything.
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.
- 7d ago Changed · +2 lines · +28 tokens per session fb28976807fc
- 16d ago Changed · +3 lines · +37 tokens per session c112043f4d6c
- 20d ago Changed · +1 lines · +11 tokens per session a221ccee788b
- 24d ago Changed · +2 lines · +38 tokens per session 09e069fd43d9
- 26d ago Changed · +2 lines · +20 tokens per session 74acb4b9a1d8
- 28d ago Changed · +2 lines · +46 tokens per session fd4e76cdf166
- 29d ago Changed · +4 lines · +96 tokens per session 89c3c81ec382
- 1mo ago Changed 72334b8affc1
- 1mo ago Changed · +4 tokens per session 82d2b11cb850
- 1mo ago First seen · 399 lines · 5,736 tokens per session scan B f94b6029507b
claude_skills AGENTS.md is an instructions file published in the GitHub repository Jamie-BitFlight/claude_skills (67 stars, last pushed 4d ago), licensed MIT. It adds 6,016 tokens to every session, about $0.0241 per session on Opus 5.5. A static security scan graded it B with 2 findings (reads agent configuration directories, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
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.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.
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).