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.
git clone --depth 1 https://github.com/paigy-ai/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/commands/paigy-ai/mcp/paigy-onboard)<a href="https://agentmods.dev/commands/paigy-ai/mcp/paigy-onboard"><img src="https://agentmods.dev/badge/commands/paigy-ai/mcp/paigy-onboard/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/commands/paigy-ai/mcp/paigy-onboard"><img src="https://agentmods.dev/badge/commands/paigy-ai/mcp/paigy-onboard.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 | $0.00012 | $0.00644 |
| Opus 5 | $0.00006 | $0.00322 |
| Sonnet 5 | $0.00002 | $0.00129 |
| Haiku 4.5 | $0.00001 | $0.00064 |
Grade B, and why
paigy-onboard 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 4d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
repo)" — and on yes, add to `~/.claude/settings.json` (or the project's What it actually says
Pair this Claude Code agent with the user's Paigy account so the Paigy MCP tools
(contact, check_replies, …) can reach them. Do it
in-context with the pair tool — no shell step needed:
- Call the
pairtool with no arguments. It attempts to open the user's browser and returns{ verification_uri_complete, user_code, device_code }. That attempt can silently fail in headless/remote environments — always show the link regardless of whether it opened. - In one line, show the user
verification_uri_completeanduser_code, and tell them to tap Approve (verifying the on-screen code matches). If the browser didn't open on its own, they can go toverification_uri_completemanually and enteruser_codethere. - In the SAME turn — do NOT stop to wait for the user — immediately call
pairagain with thatdevice_code. The tool itself polls ~90s for the approval; that is the waiting. On{ status: "pending" }, call it AGAIN with the samedevice_code(each call waits another ~90s). Keep looping until it returns the token. Only stop and ask the user if it's still pending after a couple windows (~3 min) — they may not have approved yet.
Do not end your turn on "tell me once you've approved" — the poll is automatic;
just keep calling pair. Once it returns the token, retry whatever you were doing.
-
After pairing succeeds, offer the permission allowlist (once): Paigy's value is reaching the user when they're NOT at the terminal, so the Paigy tools must never stall an unattended session on a permission prompt. Ask one question — "Allow Paigy's tools to run without prompting? (recommended; I'll add them to your Claude settings — say 'project' to scope it to this repo)" — and on yes, add to
~/.claude/settings.json(or the project's.claude/settings.jsonif they said project) underpermissions.allow:"mcp__paigy__notify_user", "mcp__paigy__check_replies", "mcp__paigy__set_task_state", "mcp__paigy__schedule_callback", "mcp__paigy__get_thread"Merge into the existing file (create keys as needed, never clobber other entries). Deliberately NOT auto-allowed:
pair/unpair— those two should stay human-approved. If they decline, drop it and don't ask again.
Fallback (older MCP without the pair tool): have the user run
npx -y @paigy/mcp@latest paigy-mcp-onboard in their shell and approve in the browser.
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.
- 4d ago Changed f5df200ca794
- 11d ago First seen · 45 lines · 12 tokens per session scan B 5a1fcf65e48c
paigy-onboard is a command published in the GitHub repository paigy-ai/mcp (0 stars, last pushed 4d ago), licensed MIT. It adds 12 tokens to every session and 644 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.