Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/alexei-led/cc-thingz/agents-mdgit clone --depth 1 https://github.com/alexei-led/cc-thingzWrote 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/alexei-led/cc-thingz/agents-md)<a href="https://agentmods.dev/instructions/alexei-led/cc-thingz/agents-md"><img src="https://agentmods.dev/badge/instructions/alexei-led/cc-thingz/agents-md.svg" alt="Measured on agentmods" 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 | $0.02377 | $0.02377 |
| Opus 5 | $0.01189 | $0.01189 |
| Sonnet 5 | $0.00475 | $0.00475 |
| Haiku 4.5 | $0.00238 | $0.00238 |
Grade A, and why
cc-thingz AGENTS.md scanned grade A with 0 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 5d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
cc-thingz — Coding Companion
Portable skills, agents, and typed hooks for Pi, Claude Code, Codex CLI, Copilot, Cursor, and Grok — code review, language tooling, infrastructure, testing, and developer utilities. Gemini is retired.
Build
make build # render package targets and typed hooks with Agent Bundler
make fmt # auto-fix ruff + shfmt + markdownlint
make check # fail on generated drift without rewriting dist
make ci # lint + validate + check + test + test-ts
make build requires an installed agbun with package, flat per-agent sidecars, declared hook environments, bundled Pi dependencies, native Pi resources, and Codex project-agent profiles. Builds need sandbox disabled because the uv cache at ~/.cache/uv is restricted in the CC sandbox.
Writing Agent/Skill Instructions
LLM signal hierarchy (MDEval benchmark + Perplexity research):
- HIGH:
#headers, bullet/numbered lists, code blocks — always use - MEDIUM:
**bold**— ≤15% of prose lines; use for bullet labels (- **Label**: desc) and critical keywords only - LOW/zero:
_italic_,---horizontal rules, markdown tables, mermaid/ASCII diagrams — never use
Specific rules:
**Label:**on its own line →### Label(real header, not bold pseudo-header)**Sentence.** followed by prose→ strip bold, keep as plain sentence---before##or**bold→ remove (redundant section break)---before````fence → keep (it's template content showing proposal format)
Run format lint: make lint-instructions or use the /reviewing-instructions skill for full scoring.
Agents
Three role agents plus one utility agent. A role is a capability envelope plus a reasoning stance no skill can supply. Domain procedure and output format live in skills; language specifics live in each skill's references/<lang>.md. Role × skill × references compose — language is not a routing key. Consolidated 39 → 3 roles (see docs/agent-audit-2026-05-16.md and the executed plan in docs/plans/completed/). runner is the cheap utility lane for simple bounded tasks, not a fourth role.
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.
- 5d ago First seen · 121 lines · 2,377 tokens per session scan A f88b397e0b8b
cc-thingz AGENTS.md is an instructions file published in the GitHub repository alexei-led/cc-thingz (35 stars, last pushed 1mo ago), licensed MIT. It adds 2,377 tokens to every session, about $0.0119 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
jko-claude-plugins CLAUDE.md
Claude Code instructions for johnkozaris/jko-claude-plugins, covering jko-skills, project structure, installing this marketplace, register as a marketplace for local development and then install any plugin from the local checkout.
jko-claude-plugins CLAUDE.md
Claude Code instructions for mathisk2095/jko-claude-plugins, covering jko-claude-plugins, project structure, installing this marketplace, register as a marketplace for local development and then install any plugin from the local checkout.
skills CLAUDE.md
Claude Code instructions for wondelai/skills, covering claude.md, repository purpose, repository structure, current skills (65) and skill file format.
agent-toolkit AGENTS.md
AGENTS.md instructions for DiUS/agent-toolkit, covering agents.md, what this repo is, repo map, principles and how to add each component type.
archeyes CLAUDE.md
Instructions for thisAAY/archeyes, covering archeyes — project guide, layout, design system, commands and publishing a release.
plumber copilot-instructions.md
Instructions for akshatnerella/plumber: You are a plumber. A plumber fixes the pipe — not the floor. The best fix makes the problem impossible.