Local-first agent platform: board + role agents + agent CLI runs (Claude Code, Cursor, Antigravity, OpenCode) or local and API models (Ollama, LM Studio), all on your own Mac
This repository also configures its own agents. See what tasktrooper tells them →
Skill Claude CodeCodex
Use when a candidate finding falls into a noisy category - the do-not-report list, the precedents that settle recurring edge cases, and the narrow exceptions where an excluded category still counts.
Skill Claude CodeCodex
Use when semgrep, osv-scanner, github-security or snyk MCP tools appear in your tool list - what each server is for, how to call it safely, what data leaves the machine, and how its findings are triaged.
Skill Claude CodeCodex
Use when deciding the verdict of a security review - severity and confidence scales, when to block, approve with notes, drop or reject as unreviewable, and the exact comment templates.
Skill Claude CodeCodex
Use when the server side of the diff fetches, renders, redirects to or validates a URL that input can influence - webhooks, link previews, import-from-URL, PDF renderers, OAuth/JWKS discovery, allowlist and blocklist checks.
Skill Claude CodeCodex
Use when the diff adds, upgrades, re-sources or re-pins a dependency - manifests, lockfiles, go.mod replace, Docker base images, GitHub Actions uses, Terraform modules, submodules, vendored code or curl-pipe-sh installs.
Skill Claude CodeCodex
Use when the diff renders user-controlled content as HTML, builds links or redirects, changes CSP or CORS, adds a cookie-authenticated state change, a postMessage handler, a markdown renderer or an Electron/Tauri window.
Skill Claude CodeCodex
Use when a screen design was approved (the design task is in done) - the hand-off document implementers build from - components mapped to the inventory, tokens, every state, responsive behaviour per breakpoint, accessibility notes and exact copy.
Skill Claude CodeCodex
Use when a design task asks you to compare a built screen with its approved design - screenshots at 1440, 768 and 375 attached to the task, Blocker/High/Medium/Nitpick triage, evidence per finding, problems not prescriptions.
Skill Claude CodeCodex
Use when you derive, change or extend a design system - where tokens live per stack, the DTCG token tree, role-based names for both themes, base vs layer vs drift, the DESIGN.md section order, the inventory format, the contrast check and the lint every proposal returns.
Skill Claude CodeCodex
Use when a screen design needs variants - how many, one document per variant, what makes two variants genuinely different, recommending one, how the human's choice is read, and narrowing to the chosen one in revision.
Skill Claude CodeCodex
Use when you draw a screen as HTML mockup documents (one per variant) - the sanitizer-safe recipe, phone and wide frames, the states grid, static control states, and the loopback self-check with screenshots at 375, 768 and 1440 attached to the task.
Skill Claude CodeCodex
Use when you write any words a user reads in a mockup or a hand-off - content language and voice, buttons, labels, errors, empty states, confirmations, numbers and dates, and the longest-string check.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: