QM is a shared work environment for coding agents that gives each employee and conversation an isolated workspace while allowing collaboration in Slack and on the web. Startups use it to run agents for individuals, teams, and company operations, with selectable harnesses and models. The catalogue entries extend QM with skills, instructions, and settings.
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 yc-software/qm --skill composiogit clone --depth 1 https://github.com/yc-software/qmWrote 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/yc-software/qm/composio)<a href="https://agentmods.dev/skills/yc-software/qm/composio"><img src="https://agentmods.dev/badge/skills/yc-software/qm/composio/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/yc-software/qm/composio"><img src="https://agentmods.dev/badge/skills/yc-software/qm/composio.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 | $0.00049 | $0.01493 |
| Opus 5.5 | $0.00020 | $0.00597 |
| Sonnet 5.5 | $0.00010 | $0.00299 |
| Haiku 4.5 | $0.00005 | $0.00149 |
Grade A, and why
composio scanned grade A 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 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- On a `502`, read the JSON body (`cause`, `providerStatus`, `message`); with Python, catch `urllib.error.HTTPError` and read `e.read()`. Back off and shrink the page rather than repeating the same request. How it starts
The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Composio
Use this skill for app discovery, consent, and execution through QM's authenticated backend. Check availability with GET /v1/composio/toolkits; an absent sandbox API key is expected. Never ask regular users to provision a Composio project key.
Show setup in web chat
When the user asks to connect apps, browse integrations, or show setup again in web chat, include the following directive as its own paragraph in your reply, with blank lines around it:
::connect-apps{}
Write the directive directly, without a code fence or quotation. The web UI renders only the searchable app picker in place. To display the separate Add to Slack action, use ::add-to-slack{} as its own paragraph instead. Include both directives in separate paragraphs when the user asks for full setup. The Slack action is available to administrators; other users should ask their administrator to install QM. Do not repeat the welcome or celebration. Rendering the widget does not authorize any service or require an SDK call; the user chooses an app and completes provider consent. Do not claim accounts are connected without verified status. For Slack conversations, use ordinary authorization links instead of this web-only directive.
When a task needs an app that isn't connected
Check connections before work that depends on an app. If the app is in the catalog but not connected, ask for it in your first reply: do any part that doesn't need the app, show the connect option, and stop. Don't finish a long partial answer first.
- Slack uses its dedicated connection flow even though
toolkitslists it. In web chat, use::link-slack-account{}as its own paragraph for personal Slack access or its connection status. Use::add-to-slack{}for the separate administrator step of installing the workspace bot, which must be completed first. In Slack conversations, direct the user to QM web Settings (/?view=settings) and Link your Slack account, signed into their existing web account. Never use::connect-apps{toolkit="slack"}or a generic Slack authorization link. These dedicated widgets do not automatically resume the task; ask the user to reply after connecting. - For other apps in web chat, put
::connect-apps{toolkit="notion"}on its own paragraph, using the exactidfromtoolkits. It shows a single connect button for that app. After consent the user returns to this conversation, the account is verified, and a message is sent so you can continue the task. - For other apps in Slack conversations, mint a link with
authorize(step 3 below) and say you'll pick the task back up when they reply. - If the app isn't in the catalog, say so in one sentence and don't show the picker. Look for the provider's official API or MCP server yourself. If it uses API keys, send a secure keychain drop link instead of asking for the key in chat.
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 43d67b12a1c1
- 3d ago Changed 53f42b18d179
- 11d ago Changed · +9 lines · +9 tokens per session 9e9ee81131d6
- 23d ago Changed · -12 lines · -6 tokens per session 119283fc5d25
- 25d ago Changed · +4 lines 1e3fad36b395
- 1mo ago Changed · +6 lines · +18 tokens per session 8f384eb7ed09
- 1mo ago First seen · 52 lines · 28 tokens per session scan A f806ed0590e0
composio is a skill published in the GitHub repository yc-software/qm (15,434 stars, last pushed today), licensed MIT. It adds 49 tokens to every session and 1,493 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-16.
Other skills, from other repositories
gog
Google Workspace CLI for Gmail, Calendar, Drive, Contacts, Sheets, and Docs.
lobster
Lobster executes multi-step workflows with approval checkpoints. Use it when.
openai-whisper-api
OpenAI Audio Transcriptions API via curl; gpt-4o-transcribe, mini, diarize, or whisper-1.
facetime
Inspect FaceTime internal stages, place an approved call to a configured owner handle, and request hangup through facetimecall.
migrating-openai-agents-sdk-to-pydantic-ai
Migrate Python OpenAI Agents SDK applications to Pydantic AI and, when warranted, Pydantic AI Harness. Use for agents.Agent, Runner, function tools, handoffs, guardrails, sessions, human approval, streaming, or SandboxAgent. Do not use for applications built directly on the OpenAI Responses API without the Agents SDK…
thisWeek
Resolve "thisWeek" to a concrete ISO date range relative to your run time — this week so far (Monday → today). Returns inclusive civil dates plus exact UTC instants so you have temporal context without computing dates by hand. Read-only: no writes, no network. Use before a week-to-date task (this week's activity…