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/jcooley8/pincushion-pluginWrote 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/jcooley8/pincushion-plugin/setup)<a href="https://agentmods.dev/commands/jcooley8/pincushion-plugin/setup"><img src="https://agentmods.dev/badge/commands/jcooley8/pincushion-plugin/setup/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/jcooley8/pincushion-plugin/setup"><img src="https://agentmods.dev/badge/commands/jcooley8/pincushion-plugin/setup.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 | $0.00013 | $0.02630 |
| Opus 5.5 | $0.00005 | $0.01052 |
| Sonnet 5.5 | $0.00003 | $0.00526 |
| Haiku 4.5 | $0.00001 | $0.00263 |
Grade A, and why
setup 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.
How it starts
The opening of the file, as written. The whole thing — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pincushion Setup
Connect this project to Pincushion so teammates and clients can pin feedback on its pages from the Pincushion app and your agent can read those pins, and compile a critique-context brief grounded in the actual page sources sitting in this repo. If the owner already created the project in the app (https://www.pincushion.io/app), pass its projectId to configure_project instead of creating a new one. The Pincushion AI critic reads this brief on every pin.
Run this once per project. You'll need:
- A project name
- The local URL where you develop (e.g.
localhost:3000) - The live/staging URL — optional but recommended. Ask for it; never guess it from the package name
- A choice of comment access mode (Pro/Team only — Free is always "open")
Steps
-
Ask the user (don't assume) which comment-access mode they want:
- Open — anyone with the review link can comment. Best for indie projects, public betas, pre-launch demos. Free, Pro, Team.
- Domain-restricted — only emails matching a list of domains. Requires Pro or Team.
- Invite-only — only emails the owner explicitly adds. Requires Pro or Team.
Separately ask: “May I inspect locally detected deployment candidates and, only after you choose one and explicitly approve, begin optional deployment automation?” Only explicit owner consent authorizes the optional
deploymentAutomation: "auto"field below; otherwise omit it. -
Call
configure_project:
configure_project({
name: "<project name>",
urls: ["localhost:3000", "myapp.vercel.app"],
commentAccess: "open" | "domain" | "invited",
allowedDomains: ["acme.com"] // only if commentAccess: "domain"
})
Save the returned projectId for step 4.
If local detection finds multiple candidates, ask the owner to select deploymentProvider; do not call a provider while it is ambiguous. If the owner explicitly authorized automation, add deploymentAutomation: "auto" (and the selected provider when needed) to that call. A successful response confirms provider read-back only; the first accepted production receipt is still pending. Unsupported provider APIs require the returned redacted manual setup handoff. Use get_deployment_reconciliation_status for independent reconciliation, fresh desktop/mobile frames, queue, and external Crit evidence. Do not modify source or CI without owner approval.
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 · 192 lines · 13 tokens per session scan A 4505e2b72cf9
setup is a command published in the GitHub repository jcooley8/pincushion-plugin (1 stars, last pushed 4d ago), licensed MIT. It adds 13 tokens to every session and 2,630 once invoked, about $0.0001 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 2026-10-23.
Other commands, from other repositories
project-proposal
Generate design proposals with scope of work, concept overview, phasing, fee structure, and terms.
spec-sheet
Draft FF&E schedules and material specifications with manufacturer details, dimensions, and lead times.
extract-screens
Analyze user stories in the current project and extract screens + reusable UI components into project-management/screens/ and project-management/components/. Requires user stories in project-management/user-stories/.
execute
Land audit batches on the audit path — fence verified first, items stay blocked until merge.
profile
Save, load, inspect, update, reset, or delete diagram-design client profiles.
setup-workflow
Set up a design team workflow and rituals.