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/cuihuan/awesome-ai-gateway/claude-mdgit clone --depth 1 https://github.com/cuihuan/awesome-ai-gatewayWhat 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 | $0.00988 | $0.00988 |
| Opus 5 | $0.00494 | $0.00494 |
| Sonnet 5 | $0.00198 | $0.00198 |
| Haiku 4.5 | $0.00099 | $0.00099 |
Grade A, and why
awesome-ai-gateway CLAUDE.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 yesterday.
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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — working rules for this repository
Read this before changing anything here. It states what the project is for, so that edits can be judged against a goal instead of taste.
What this repository is
awesome-ai-gateway is not a link dump. It is meant to be the integration point for the
AI-gateway space: one place where a visitor can see the whole landscape — open-source
projects, hosted services, relays, the protocols they translate, and the evidence behind
every claim — without having to trust a vendor.
The two goals — judge every change against these
Goal 1 — a visitor understands the space at a glance. Within the first screen and a few clicks, a reader should get three things:
- The state of play — what exists, what consolidated, what died, what is actually maintained. Not a catalogue: a picture.
- How to choose — the decision path from their requirement to a specific name, with the trade-off stated.
- Where it is heading — the direction of travel, with dated evidence and a stated way to be proven wrong.
If a change does not make one of those three clearer, it is decoration. Length is not value: a section that makes the first screen slower to read is a net loss even if every sentence in it is true.
Goal 2 — grow stars and users, because that is the honest signal of Goal 1. Stars are not vanity here; they are the proxy for "this actually helped someone choose." So optimise for the things that earn a star from an engineer in a hurry:
- A reusable artifact beats a list entry (a rubric, a computed table, a reproducible test, a calculator). People star what they will come back to.
- A number nobody else publishes beats a paraphrased vendor claim.
- Being right when it is inconvenient — naming an archived project, a CVE, a stale vendor promise — is what makes the rest of the list credible.
Never chase the metric directly: no star-for-star, no reciprocal-listing deals, no inflating entry counts. Those raise the number and destroy the reason it means 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.
- yesterday First seen · 82 lines · 988 tokens per session scan A 930269e8ab49
awesome-ai-gateway CLAUDE.md is an instructions file published in the GitHub repository cuihuan/awesome-ai-gateway (97 stars, last pushed yesterday), licensed CC0-1.0. It adds 988 tokens to every session, about $0.0049 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
deltallm AGENTS.md
Instructions for deltawi/deltallm, covering repository instructions, required first step and precedence and maintenance.
awesome-go AGENTS.md
Instructions for avelino/awesome-go, covering awesome-go · llm contribution guide, project snapshot, when modifying the awesome list, coding guidelines and testing & validation.
higress AGENTS.md
AGENTS.md instructions for higress-group/higress, covering agents.md, mandatory issue-spec gate for agent-assisted changes, repository layout, plugins and plugins/wasm-go/ (primary wasm plugin framework, go).
awesome-ChatGPT-repositories CLAUDE.md
Instructions for taishi-i/awesome-ChatGPT-repositories, covering awesome-chatgpt-repositories — claude code guide, repository structure, plugin skill (when installed via /plugin), local standalone command (when repo is cloned) and compact data format (plugins/awesome-chatgpt-search/data/).
awesome-vscode CLAUDE.md
Instructions for viatsko/awesome-vscode, covering claude.md, what this repo is, adding an entry, themes and linting / ci.
awesome-python AGENTS.md
AGENTS.md instructions for vinta/awesome-python, covering agents.md and entry rules.