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 AiAi-Mastermind/aiai-mastermind-tools-and-skills --skill muse-browsergit clone --depth 1 https://github.com/AiAi-Mastermind/aiai-mastermind-tools-and-skillsWrote 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/aiai-mastermind/aiai-mastermind-tools-and-skills/muse-browser)<a href="https://agentmods.dev/skills/aiai-mastermind/aiai-mastermind-tools-and-skills/muse-browser"><img src="https://agentmods.dev/badge/skills/aiai-mastermind/aiai-mastermind-tools-and-skills/muse-browser/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/aiai-mastermind/aiai-mastermind-tools-and-skills/muse-browser"><img src="https://agentmods.dev/badge/skills/aiai-mastermind/aiai-mastermind-tools-and-skills/muse-browser.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 | $0.00159 | $0.01726 |
| Opus 5.5 | $0.00064 | $0.00690 |
| Sonnet 5.5 | $0.00032 | $0.00345 |
| Haiku 4.5 | $0.00016 | $0.00173 |
Grade B, and why
muse-browser scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Enumerates other installed skillsmediumAgent snooping
Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.
MA="$(ls -d ~/.claude/skills/muse-browser ~/.codex/skills/muse-browser ~/.agents/skills/muse-browser 2>/dev/null | head -1)/bin/muse-ask" How it starts
The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Muse (Meta's AI agent) for Facebook and Instagram research
EXPERIMENTAL. On 2026-10-07 this driver's side-chat isolation check failed in testing: it could not positively prove it had landed in a fresh, empty side chat, so it refused to send. The refusal is the safe outcome, but it means this lane may return
isolation_unavailableuntil the check is updated for Muse's current interface. Treat any result as a lead, check the isolation proofs yourself when driving by hand, and never send a question into your main Muse chat to get around it.
Muse is Meta's personal AI agent at https://muse.ai/. It is not meta.ai, which is Meta's separate chat. Muse signs in through your Facebook account and has first-party search over Facebook, Instagram and Threads posts your account can see, including Facebook groups you belong to. Asking Muse is far safer for your account than letting an agent scroll Facebook, because it is Meta's own product doing the searching, not automated scraping of the site.
Requires the aiai-agent-chrome kit, installed and running, with you signed in to Facebook and muse.ai inside the Agent Chrome window.
Use the driver
MA="$(ls -d ~/.claude/skills/muse-browser ~/.codex/skills/muse-browser ~/.agents/skills/muse-browser 2>/dev/null | head -1)/bin/muse-ask"
"$MA" "What are Facebook groups and Instagram creators saying in the last 30 days about <topic>?" --json
A run usually takes 1-2 minutes (up to a 10-minute timeout, then partial). The output is one JSON object with status, answer_markdown, sources (instagram.com/reel and facebook.com/groups/.../permalink links), elapsed_s and file (a private copy under ~/Library/Application Support/BrowserResearch/runs/<date>/).
| status | do this |
|---|---|
done |
use it; verify the posts you lean on (below) |
partial |
Muse was still working at the timeout; the answer is incomplete. Use it only as a partial lead and say so |
signin_needed |
tell the owner to sign in to Facebook in Agent Chrome (Muse signs in through it): "$AC" open https://muse.ai/ |
isolation_unavailable |
the driver didn't land in an empty side chat, so nothing was sent. Retry once. Never send into the main chat. See the EXPERIMENTAL note above |
needs_approval |
Muse wanted to act. Nothing was approved; report it |
rate_limited |
Muse's usage pool is spent; stop |
ui_changed / timeout |
report it, or drive it yourself (below) |
not_installed / not_running / error |
the kit is missing or Agent Chrome is down: "$AC" doctor, report the message |
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 70 lines · 159 tokens per session scan B 496437fe53e9
muse-browser is a skill published in the GitHub repository AiAi-Mastermind/aiai-mastermind-tools-and-skills (2 stars, last pushed 4d ago), licensed MIT. It adds 159 tokens to every session and 1,726 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it B with 1 finding (enumerates other installed skills). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-09.
Other skills, from other repositories
defuddle
Plan and, with explicit network consent, use an optional external Defuddle cleaner to extract article-like HTTPS pages as Markdown. Use for defuddle, clean this URL, strip page clutter, readable Markdown from a web page, or preparing a web source for later wiki ingestion.
apply
A browser-based job application workflow for submitting applications through services such as Ashby, Greenhouse, Lever, and Hacker News. It uses prepared candidate information, answers, job links, and cover letters.
dr-fanout
Fan one deep-research prompt out to several external LLMs at once through a logged-in browser, then collect the reports, archive the originals and synthesize a consensus. Use at the final stage of a research protocol, instead of hand-pasting one prompt into N sites. Triggers: "/dr-fanout", "fan out the DR"…
cv-outreach
A daily workflow for carrying out scheduled outward-facing tasks, such as event registrations, messages to organisers, and public posts. It uses a browser identity carefully so actions are made from the intended account.
social-daily
A workflow for publishing prepared content to Facebook, X, Telegram, Threads, and Instagram. It uses a shared queue so posts for different platforms can be tracked together.
fb-reply
Read who commented on recent Facebook posts and publish personalized replies through the real logged-in Chrome tab, draft-first, with a daily cap and a minimum gap between replies. Triggers: "/fb-reply", "reply to my facebook comments".