Borrowing it
Nothing to install: this file belongs to thisnick/agent-wechat. 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/thisnick/agent-wechat/main/AGENTS.mdgit clone --depth 1 https://github.com/thisnick/agent-wechatWrote 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/thisnick/agent-wechat/agents-md)<a href="https://agentmods.dev/instructions/thisnick/agent-wechat/agents-md"><img src="https://agentmods.dev/badge/instructions/thisnick/agent-wechat/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/thisnick/agent-wechat/agents-md"><img src="https://agentmods.dev/badge/instructions/thisnick/agent-wechat/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 | $0.02612 | $0.02612 |
| Opus 5.5 | $0.01045 | $0.01045 |
| Sonnet 5.5 | $0.00522 | $0.00522 |
| Haiku 4.5 | $0.00261 | $0.00261 |
Grade A, and why
agent-wechat 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 4d 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 — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent guide for agent-wechat
This repository provides a standalone native WeChat daemon, Rust CLI and OpenClaw channel. It communicates directly with WeChat and reads/writes supported official-format profiles without an official executable at runtime. Check implementation and validation before describing a capability as supported.
Repository map
| Path | Responsibility |
|---|---|
packages/native-client-rust/ |
Protocol, profile, sidecar store, account actors, HTTP daemon and Rust wx CLI |
packages/shared/ |
Private native TypeScript HTTP/SSE SDK and generated Rust contracts |
packages/cli/ |
npm wx launcher and verified executable installer |
packages/openclaw-extension/ |
OpenClaw native hosting, event custody, model handoff and media |
deploy/native/ |
Native-only image, Compose and optional user-service examples |
docs/ |
Public documentation site |
Native routes/contracts live in packages/native-client-rust/crates/wechat-api/; run pnpm generate-types after exported Rust contract changes. The README is a concise user guide, not an exhaustive API specification. The former UI server, image, container CLI and Wechaty integrations are retired; Git history retains their implementation.
Working rules
- Use neutral, accurate technical language in code, comments, docs, commits, and PRs. Do not imply unauthorized access or conceal how a feature works.
- Keep account credentials, tokens, private infrastructure details, and research-only material out of this public repository. Do not link to private research repositories from public docs.
- WeChat protocol, profile-format and reverse-engineering research findings are not kept in this repository. If a dedicated internal research/documentation repository is available, read and record those findings there instead; do not add
dev-notes/-style research or validation notes here. - Use an isolated test container and test account for behavior changes. Do not send messages to a live bot or modify a production container unless the task explicitly calls for it. Confirm the intended destination before any send.
- Preserve existing container data, official profiles and backups unless removal is explicitly requested. Retiring runtime source does not authorize deleting deployed containers, registry tags or volumes.
- Use one writer per official profile. Stop the official app before takeover; a caller assertion is not proof that an unmanaged app stopped. Do not infer arbitrary official-build or live ARM64 compatibility from synthetic tests.
- Add a changeset for user-facing behavior or package changes. Documentation-only edits do not need one.
- Follow the official client's verified implementation and storage for existing WeChat features. Settled message/contact/group state belongs in the supported official profile; read or derive it there instead of creating a parallel settled-state model in a sidecar. Sidecars may track pending work, raw receipts, deduplication, idempotency and recovery. Keep those operational records separate from canonical state and document why each is needed. An API may derive additional information from existing evidence without persisting a second copy. Treat behavior or retention beyond the official client as a separate proposed feature, state that difference, and obtain user agreement before implementing it. Missing or incomplete analysis is not evidence that extra storage is necessary.
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.
- 4d ago Changed · +5 lines · +54 tokens per session b1bf04566006
- 13d ago Changed · +1 lines · +71 tokens per session 2d7b407b3b08
- 14d ago Changed · +1 lines · +136 tokens per session 3c2981cc9c31
- 15d ago Changed · +58 lines · +1,153 tokens per session 770059d9bc5f
- 16d ago Changed · -14 lines · -649 tokens per session 17c809a6b3d4
- 29d ago Changed · +2 lines · +172 tokens per session bb3307e3e0b5
- 1mo ago Changed · +67 tokens per session f9a40b755348
- 1mo ago Changed · +2 lines · +97 tokens per session 3d33bb6e02f5
- 1mo ago First seen · 69 lines · 1,511 tokens per session scan A e2a70b8dd61d
agent-wechat AGENTS.md is an instructions file published in the GitHub repository thisnick/agent-wechat (245 stars, last pushed yesterday), licensed MIT. It adds 2,612 tokens to every session, about $0.0104 per session on Opus 5.5. 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 2026-09-24.
agent-wechat AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| agent-wechat AGENTS.mdthisnick/agent-wechat | Instructions file | 245 | yesterday | MIT | A | 2,612 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |