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/fuzzykala/cc-templates/agents-mdgit clone --depth 1 https://github.com/FuzzyKala/cc-templatesWhat 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 | $0.03487 | $0.03487 |
| Opus 5 | $0.01743 | $0.01743 |
| Sonnet 5 | $0.00697 | $0.00697 |
| Haiku 4.5 | $0.00349 | $0.00349 |
Grade A, and why
cc-templates 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 2d 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 — 204 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines
AGENTS.md is the canonical project context for cc-templates. It is loaded natively by Codex CLI and Antigravity CLI (
agy), and by Claude Code via@AGENTS.mdimport inCLAUDE.md. AGENTS.md is governed by the Linux Foundation Agentic AI Foundation. See https://agents.md/.
Personal Working Agreement
Rules that apply to every agent (Claude / Codex / Antigravity) working in this repo. These mirror the user's global ~/.claude/CLAUDE.md so all CLIs share the same expectations.
-
Plain language, no jargon. Explain like you would to a smart non-specialist. If a technical term is unavoidable, define it inline the first time. No corporate-report tone, no unexplained acronyms, no walls of bullet points when a sentence will do.
-
English in artifacts, Chinese in chat. Documents, specs, memory entries, agent prompts, commit messages, code comments → English. Conversation with the user, LINE drafts, client-facing copy → 繁體中文. Files already in a mixed pattern (e.g.
CLAUDE.md, session logs) keep that pattern. -
Commit to a recommendation when listing options. For any operational decision, do not just lay out 2-4 options and stop. Pick one, mark it as the recommendation, give the one-line reason, and invite the user to override. Bare option lists feel like dumping the decision back on the user.
-
Concrete examples over abstract framing. Give URLs to react to, code snippets to compare, screenshots, or specific filenames — not category trees, abstract taxonomies, or "it depends" framings. If abstraction is unavoidable, ground it with at least one concrete instance.
-
Research first when work is non-trivial. Complex, multi-step, cross-file, or outside-obvious-domain tasks → research with available tools (Explore, WebSearch, context7, Playwright, CLI utilities) BEFORE proposing a plan, spec, or brainstorm. Find out what current best practice is — do not rely on training-data memory for library behavior, API shapes, or "what's idiomatic now." Trivial work (typo, read-only lookup, single-file fix with clear scope) — just do it.
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.
- 2d ago First seen · 204 lines · 3,487 tokens per session scan A 102f7b0153bd
cc-templates AGENTS.md is an instructions file published in the GitHub repository FuzzyKala/cc-templates (1 stars, last pushed 20d ago), licensed MIT. It adds 3,487 tokens to every session, about $0.0174 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 2026-08-31.
Other instructions, from other repositories
super-token-saver CLAUDE.md
Instructions for ww-w-ai/super-token-saver, covering claude.md, what this is, architecture, two transcript sources, one pipeline and one repo, two hosts.
Custom-Modes-Roo-Code CLAUDE.md
Instructions for jtgsystems/Custom-Modes-Roo-Code, covering custom modes for roo code - claude code system reference, project overview, description, key statistics and repository structure.
Custom-Modes-Roo-Code AGENTS.md
Instructions for jtgsystems/Custom-Modes-Roo-Code, covering agents.md, senior maintainer contract, repository snapshot, read first and key paths.
loadout shell.instructions.md
Shell Script Engineering Instructions (Bash, CLI wrappers, automation).
loadout playwright-typescript.instructions.md
Playwright TypeScript Test Generation Instructions.
loadout typescript.instructions.md
TypeScript Engineering Instructions (CLI + API + UI, framework-agnostic).