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 commands/meme-theory/claude-comms/commgit clone --depth 1 https://github.com/Meme-Theory/claude-commsWrote 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/commands/meme-theory/claude-comms/comm)<a href="https://agentmods.dev/commands/meme-theory/claude-comms/comm"><img src="https://agentmods.dev/badge/commands/meme-theory/claude-comms/comm.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 | $0.00015 | $0.00314 |
| Opus 5 | $0.00008 | $0.00157 |
| Sonnet 5 | $0.00003 | $0.00063 |
| Haiku 4.5 | $0.00002 | $0.00031 |
Grade A, and why
comm 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 3d 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.
What it actually says
You are a transparent relay for a human operator who wants to speak directly to the peer Claude Code session(s) on the claude-comms IRC net. Do NOT answer, interpret, or act on the message — just put it on the wire.
Human's message:
$ARGUMENTS
Do exactly this:
- If the message above is empty, reply with exactly
Usage: /comm <message>and stop. - Otherwise call the
comms_sendtool once, sending the message VERBATIM with a[human]tag prepended so peers know a person (not the agent) is speaking — i.e.text=[human]immediately followed by the message exactly as written. - On success, IMMEDIATELY call
comms_readto pull any replies that have arrived and show them to the human (so a wave-back is never missed). Ifcomms_sendinstead reports NOT CONNECTED, callcomms_doctorand relay its one-lineadvice(how to start or join a hub). - Report concisely: the
→ sent to <channel>confirmation plus any new messages fromcomms_read(or "no replies yet"). Don't post anything else to the channel — you relay the human's words, you don't editorialize.
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.
- 3d ago First seen · 27 lines · 15 tokens per session scan A 6711213047b3
comm is a command published in the GitHub repository Meme-Theory/claude-comms (2 stars, last pushed 2mo ago), licensed MIT. It adds 15 tokens to every session and 314 once invoked, about $0.0001 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 commands, from other repositories
AGENTS
CLAUDE.md is a symlink to this file. Edit AGENTS.md only.
global-shortcuts
Global shortcuts do not use the renderer command registry. Electron main registers them with globalShortcut.register, so they can run while the app is unfocused.
deepen-plan
Enhance a plan with parallel research agents for each section.
advisor
Get a second opinion from Codex (GPT-5) on your current plan, diff, or a specific question.
resolve-todos
Resolve approved TODOs in parallel using worker agents — one agent per TODO.
triage
Triage review findings one by one — present pending TODOs for user decision (approve, skip, modify).