Fork a working agent template. Skip the setup.

Margin & Pricing Check

Claude Code

$25/mo

True margin from landed cost — not markup, not invoice cost — plus price scenarios with the break-even volume change. Never changes prices.

Payroll Liability Reconciler

Claude Code

$25/mo

Ties gross-to-net per period and catches the employer-vs-employee coding errors that hide in a payroll liability balance. Not tax advice; never posts.

sentry-incident-brief-sf

Claude Code

$25/mo

Error cluster → timeline, suspect commits, draft postmortem. You page. Never auto-mute or auto-deploy.

ci-green-keeper-sf

Claude Code

$25/mo

When main goes red, draft the smallest green fix PR. Human merge gate. Never force-push main.

CodeCrucible AI Vulnerability Scanner

Claude Code

$25/mo

AI vulnerability scanner / AI SAST on Claude Code: Block CodeCrucible, dry-run cost gate, SARIF out. You verify findings — nothing auto-merges.

Postiz Social Media Scheduler

OpenClaw

$25/mo

Postiz + OpenClaw: tell one agent what to post and when. It drafts and queues across the 28 channels Postiz lists. Bring a Postiz API key.

vLLM Inference Lab: p99 Bench

OpenScience

$49/mo

OpenScience lab for OpenAI-compatible inference: design a vLLM-style serve stack, bench it, read real p50/p99. BYOK Anthropic plus optional HF token.

AI Co-Scientist for ML Research

OpenScience

$49/mo

OpenScience co-scientist for ML: literature loop, one concrete experiment, train/eval scaffold, methods draft — as files. Bring your own Anthropic key.

AI Drug Discovery Agent (UniProt/PDB)

OpenScience

$49/mo

Target briefs for AI drug discovery, grounded in UniProt, PDB and ChEMBL accessions you can click and check. Bring one model key. Nothing docked or predicted.

Flaky Test Hunter

Claude Code

$25/mo

Flaky test hunter for GitHub Actions: groups CI failures by test, scores flake vs real bug, drafts the quarantine or fix PR. You merge every change.