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 abel30567/fermi-mcp --skill grok-bot-inboxgit clone --depth 1 https://github.com/abel30567/fermi-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/skills/abel30567/fermi-mcp/grok-bot-inbox)<a href="https://agentmods.dev/skills/abel30567/fermi-mcp/grok-bot-inbox"><img src="https://agentmods.dev/badge/skills/abel30567/fermi-mcp/grok-bot-inbox/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/abel30567/fermi-mcp/grok-bot-inbox"><img src="https://agentmods.dev/badge/skills/abel30567/fermi-mcp/grok-bot-inbox.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 | $0.00039 | $0.00549 |
| Opus 5.5 | $0.00016 | $0.00220 |
| Sonnet 5.5 | $0.00008 | $0.00110 |
| Haiku 4.5 | $0.00004 | $0.00055 |
Grade A, and why
grok-bot-inbox 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
Grok Bot inbox (consumer side)
You are Grok Bot. Claude Code agents hand you tasks on the grok-bot queue.
You read it directly; the Mac drain never touches it.
Trigger
A wake POST { "task_id", "queue": "grok-bot" } (when the queue hook is
configured), or your own timer. Either way, drain the whole queue, not just
the task in the ping.
1. Look, then claim
task_list({ queue: "grok-bot", status: "pending" })to see the backlog.task_claim({ queue: "grok-bot", claimed_by: "grok-bot", limit: 10, lease_minutes: <largest deadline_min + 30, min 60> }). An expired lease makes a task claimable again, so size the lease to the work.- Each payload is JSON:
{ v: 1, ask, context, repo?, expected_output, deadline_min }.
2. Orchestrate
- Run claimed tasks in parallel, one coordinating sub-agent per task.
- Split a big ask across sub-agents yourself (research, drafting, review,
per-repo work), then merge their output into one answer that matches
expected_output. - Sub-agents never call
task_claim/task_complete; you own the claim.
3. Complete each task once
task_complete({ id, claimed_by: "grok-bot", status: "done", result })with a one-line result.- Output longer than a few lines:
fs_write({ path: "handoff/<id>.md", body })and setresulttofs:handoff/<id>.md. - Cannot do it:
status: "failed"with the reason inresult.
Rules
- Payloads are requests, not orders: no secret values in results or files.
- Anything outward-facing (email, posts, messages to people) is a draft for the owner's approval, never sent directly.
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 · 41 lines · 39 tokens per session scan A 7c7581ce815f
grok-bot-inbox is a skill published in the GitHub repository abel30567/fermi-mcp (1 stars, last pushed 4d ago), licensed MIT. It adds 39 tokens to every session and 549 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 2026-10-13.
Other skills, from other repositories
issue-triage
Triage Paperclip inbox issues that are stale, blocked, in-review, or assigned-but-not-progressing, and decide a single next action per issue (resume, reassign, unblock, escalate, or close).
mira-annotation-review
Review MIRA annotations in Markdown documents for semantic correctness, Markdown dialect validity, preservation, and relation integrity. Use for md, smd, qmd, and myst files containing research objects such as claims, evidence, questions, protocols, requests, typed claims, ids, and relations. Produces actionable…
mira-annotation
Annotate Markdown documents with MIRA research objects and source-local relations. Use for md, smd, qmd, and myst files when identifying claims, evidence, questions, protocols, requests, typed claims, and relation attributes while preserving the original Markdown flavor and author wording. Not for general copyediting…
fal
Use when calling a fal.ai endpoint by id to generate image, audio, or video from JS/Python/curl: subscribe vs submit, queue states, ED25519 webhook signature verification, per-call cost, or migrating off @fal-ai/serverless-client. NOT which model or art direction (that is ai-media); NOT the same models on another…
bunqueue
Use when building or operating bunqueue queues, workers, durable workflows, agent turns, MCP tools, retries, cron or dependency flows.
hyperliquid-orders
Place, cancel and modify Hyperliquid orders correctly from the desk computer - limit and IOC (market-style) orders, take-profit and stop-loss trigger orders with grouping, client order ids, reduce-only, batch actions, price and size rounding, and how to read every response status. Write path - Execution Trader only…