Borrowing it
Nothing to install: this file belongs to aaronsb/slack-mcp. 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/aaronsb/slack-mcp/main/CLAUDE.mdgit clone --depth 1 https://github.com/aaronsb/slack-mcpWrote 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/aaronsb/slack-mcp/claude-md)<a href="https://agentmods.dev/instructions/aaronsb/slack-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/aaronsb/slack-mcp/claude-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/aaronsb/slack-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/aaronsb/slack-mcp/claude-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 | $0.01869 | $0.01869 |
| Opus 5.5 | $0.00748 | $0.00748 |
| Sonnet 5.5 | $0.00374 | $0.00374 |
| Haiku 4.5 | $0.00187 | $0.00187 |
Grade A, and why
slack-mcp 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 17d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Slack MCP Server
Go-based MCP server for Slack workspace interaction using session tokens (xoxc/xoxd). No OAuth, no bot permissions, no admin approval.
Build
make build # Build for current platform
make build-all-platforms # Cross-compile (darwin/linux/windows x amd64/arm64)
make test # Run tests
make format # Format code
make tidy # Tidy go modules
make clean # Remove build artifacts
make npm-publish NPM_TOKEN=... # Publish to npm
Architecture
cmd/slack-mcp/— Entry point, transport selection (stdio/sse), setup commandpkg/server/— MCP server, tool registrationpkg/provider/— Slack API client, two-phase channel cachingpkg/features/— Tool implementationspkg/text/— Text processing utilitiesnpm/— npm wrapper packages (platform binary resolver)
Tools
| Tool | Kind | What it does |
|---|---|---|
inbox |
noun | What needs you: view='new' (since your last dismiss), 'unreads', 'mentions' |
messages |
noun | Conversation content: target= reads in full, +around= context, +since= time window, query= raw Slack search syntax (passed as written) plus resolved filters in, from, after, before, has (link, pin, :emoji:), thread; scheduled=true lists your pending scheduled messages, soonest first, and never your unsent composer drafts (ADR-016) |
estate |
noun | Workspace shape and relationships: view='about'|'families'|'person'|'initiatives'|'convergence'|'people'|'channels'; about/person take render='graph' to also write a static HTML graph page (not in batch) |
batch |
executor | Run a held plan of reads in one call: commands=[{tool, params}...]; playbooks via save=/run=/list=/delete= |
say |
verb | Contribute content (Slack-visible): a message (a thread reply can also go to the channel with broadcast=true), files from the exchange directory (files=['report.pdf'], bare names, at most 10, shared as one message with text as the comment), or an emoji reaction; at= schedules a message for Slack to send later (2 minutes to 120 days; a time with no offset is read in your Slack profile zone and refused if this machine's zone disagrees) and cancel=<handle> withdraws one (ADR-016); scanned for secrets and gated before anything is sent or scheduled, with the send time part of what an approval covers (ADR-013) |
dismiss |
verb | Mark inbox items handled — private watermark, invisible to Slack |
mark-read |
verb | Fire read receipts — the one visibly-public read signal; refused at a quarantined destination, and never opens a DM |
auth |
verb | Interactive token setup (localhost only) |
download |
verb | Download a shared file into the exchange directory (filename= is a bare name, not a path; a taken name is saved as name (n).ext and the result says so) |
put |
verb | Write a new file into the exchange directory for say files= to attach, for clients whose file tools cannot reach the server's filesystem: name= (bare name) and exactly one of content= (text) or base64= (standard alphabet, padding optional); at most 5 MiB; never overwrites (a taken name is saved as name (n).ext), never reads, lists, or deletes (ADR-012) |
unlock |
verb | Open a local page (127.0.0.1, one-shot link, sent to the browser only) where the operator reviews and clears quarantines and the strike lock; the tool clears nothing, returns no link, and never reports what was cleared; refused on remote and SSE (ADR-013, #132) |
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.
- 17d ago Changed · +1 lines · +147 tokens per session f5100ce2e0f4
- 23d ago Changed · +7 lines · +1,048 tokens per session 2e9d6449a2ae
- 1mo ago First seen · 54 lines · 674 tokens per session scan A 7a7e128d637b
slack-mcp CLAUDE.md is an instructions file published in the GitHub repository aaronsb/slack-mcp (2 stars, last pushed 4d ago), licensed MIT. It adds 1,869 tokens to every session, about $0.0075 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 2026-08-31.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.