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 skills add mindroom-ai/mindroom --skill agent-conversationsgit clone --depth 1 https://github.com/mindroom-ai/mindroomWrote 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/skills/mindroom-ai/mindroom/agent-conversations)<a href="https://agentmods.dev/skills/mindroom-ai/mindroom/agent-conversations"><img src="https://agentmods.dev/badge/skills/mindroom-ai/mindroom/agent-conversations/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/skills/mindroom-ai/mindroom/agent-conversations"><img src="https://agentmods.dev/badge/skills/mindroom-ai/mindroom/agent-conversations.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 | $0.00054 | $0.01049 |
| Opus 5.5 | $0.00022 | $0.00420 |
| Sonnet 5.5 | $0.00011 | $0.00210 |
| Haiku 4.5 | $0.00005 | $0.00105 |
Grade A, and why
agent-conversations 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 today.
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 — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent conversations
An agent takes a turn only when a message wakes it.
Your reply is posted in this conversation for everyone to read, but it wakes only the agents it mentions.
Call matrix_message only to wake an agent, so every call has a recipient, unless the person asks you to send a file, post in another room or thread, or read, edit, or react to a message; everything else you say goes in your reply.
Without the matrix_message tool you still reach agents through mentions in your reply, as the last section shows.
What wakes an agent
| You do | Effect |
|---|---|
matrix_message(recipient="critic", message="..."), with an agent name from matrix_room(action="agents") |
That agent takes a turn in this conversation, or on the room timeline when its thread_mode is room; the call returns at once, without its answer |
matrix_message without recipient |
Posts a message that wakes no one, even if it contains @names |
@critic or the agent's Matrix ID in your reply |
That agent takes a turn after your reply finishes |
| Two or more agents mentioned in one reply, counting yourself | They answer together as one team reply, or nobody answers |
A display name such as "the Devil's Advocate", without @ |
Nobody wakes |
A message from an agent wakes you only if it mentions you or names you as recipient.
Asking an agent
- Call
matrix_message(recipient="<name>", message="<request>")once; without that tool, ask in your reply as the last section says. Write a self-contained request and end it with one of these sentences, filling in your ownname, or your full Matrix ID in place of@<your name>:- you need the answer to finish: "When done, start your reply with @."
- the person should get the answer: "Reply with the answer for the person and mention no agent."
- you only inform it, even when the person says it need not reply: "Reply with one short line and mention no agent." Every agent you wake must reply, so never tell it not to.
- End your turn with your reply: one or two sentences telling the person what you asked, mentioning no agent. A hand-back, a reply that starts with your mention, wakes you in a new turn, so do not read the conversation to wait for it.
- When the hand-back wakes you, continue the task. If your earlier reply already used that answer, reply in one short line that mentions no agent.
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.
- today Changed · +8 lines 62e40a12d31e
- yesterday First seen · 60 lines · 54 tokens per session scan A d4d7a4962ec4
agent-conversations is a skill published in the GitHub repository mindroom-ai/mindroom (335 stars, last pushed today), licensed Apache-2.0. It adds 54 tokens to every session and 1,049 once invoked, about $0.0002 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 2026-10-15.
Other skills, from other repositories
ctf-pwn
Solve CTF binary exploitation challenges by discovering and exploiting memory corruption vulnerabilities to read flags. Use for buffer overflows, format strings, heap exploits, ROP challenges, or any pwn/exploitation task.
ctf-rev
Solve CTF reverse engineering challenges using systematic analysis to find flags, keys, or passwords. Use for crackmes, binary bombs, key validators, obfuscated code, algorithm recovery, or any challenge requiring program comprehension to extract hidden information.
deep-analysis
Performs focused, depth-first investigation of specific reverse engineering questions through iterative analysis and database improvement. Answers questions like "What does this function do?", "Does this use crypto?", "What's the C2 address?", "Fix types in this function". Makes incremental improvements (renaming…
pyghidra-scripting
Write and run Python (PyGhidra) code inside the Ghidra session that ReVa's MCP server is already attached to, using the five ReVa scripting tools — run-script, list-scripts, read-script, write-script, edit-script. Use this whenever the user asks to execute Python against the current program, reach for the Ghidra Flat…
ctf-crypto
Solve CTF cryptography challenges by identifying, analyzing, and exploiting weak crypto implementations in binaries to extract keys or decrypt data. Use for custom ciphers, weak crypto, key extraction, or algorithm identification.
gmail-email
Send emails via Gmail SMTP using App Password authentication.