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/anneschuth/claude-threads/claude-mdgit clone --depth 1 https://github.com/anneschuth/claude-threadsWrote 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/anneschuth/claude-threads/claude-md)<a href="https://agentmods.dev/instructions/anneschuth/claude-threads/claude-md"><img src="https://agentmods.dev/badge/instructions/anneschuth/claude-threads/claude-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 | $0.14337 | $0.14337 |
| Opus 5 | $0.07168 | $0.07168 |
| Sonnet 5 | $0.02867 | $0.02867 |
| Haiku 4.5 | $0.01434 | $0.01434 |
Grade B, and why
claude-threads CLAUDE.md scanned grade B with 1 finding 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
tail -20 ~/.claude/history.jsonl | jq -r '.cwd + " " + .name' How it starts
The opening of the file, as written. The whole thing — 1,100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude Code Instructions for claude-threads
What This Project Does
This is a multi-platform bot that lets users interact with Claude Code through chat platforms. When someone @mentions the bot in a channel, it spawns a Claude Code CLI session in a configured working directory and streams all output to a thread. The user can continue the conversation by replying in the thread.
Currently Supported Platforms:
- Mattermost (full support)
- Slack (full support)
Key Features:
- Real-time streaming of Claude responses to chat platforms
- Multi-platform support - connect to multiple Mattermost/Slack instances simultaneously
- Multiple concurrent sessions - one per thread, across all platforms
- Session persistence - sessions resume automatically after bot restart
- Session collaboration -
!invite @userto temporarily allow users in a session - Message approval - unauthorized users can request approval for their messages
- Thread context prompt - when starting a session mid-thread, offers to include previous conversation context
- Interactive permission approval via emoji reactions
- Plan approval and question answering via reactions
- Task list display with live updates
- Code diffs and file previews
- Multi-user access control
- Automatic idle session cleanup
- Permalink follower (
read_postMCP tool) - Claude can resolve a Mattermost or Slack permalink to its content (and optional thread context) inside the bot's own channel - Persistent memory - per-channel shared notes (
!remember/!memory, Claude Tag style) plus Claude Code's native auto-memory redirected into bot-managed per-(platform, repo) directories; end-of-session distillation learns team facts over time - Routines - scheduled recurring work (
!routine every weekday at 9am, ..., Claude Tag style): natural-language creation with 👍 confirmation, fired as bot-initiated session threads;!routinesto list/pause/resume/delete/run - Watches - event triggers (
!watch when someone reports an incident, ...): a free keyword prefilter plus a haiku semantic confirm fire a session in the triggering message's own thread;!watchesto list/pause/resume/delete
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 · 1,100 lines · 14,337 tokens per session scan B 5715e3c3c2da
claude-threads CLAUDE.md is an instructions file published in the GitHub repository anneschuth/claude-threads (39 stars, last pushed 2d ago), licensed Apache-2.0. It adds 14,337 tokens to every session, about $0.0717 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
ccxt CLAUDE.md
Claude Code instructions for ccxt/ccxt, covering ccxt — repository guide for contributors and ai agents, 1. architecture, rest vs pro (websocket), per-language model and two transpilers.
nanobot AGENTS.md
AGENTS.md instructions for HKUDS/nanobot, covering project overview, development commands, python: run single test / lint, strict type checking (matches ci) and gateway.
clawsweeper AGENTS.md
AGENTS.md instructions for openclaw/clawsweeper, covering agents.md, structure, operating model, safety rules and pr validation and landing.
clawpatch AGENTS.md
AGENTS.md instructions for openclaw/clawpatch, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
Undefined CLAUDE.md
Instructions for 69gg/Undefined, covering claude.md, 项目概述, 开发命令, 依赖安装 / 初始化 and 启动.
yutu AGENTS.md
AGENTS.md instructions for eat-pray-ai/yutu, covering yutu, quick reference, directory index, documentation and conventions.