Guide for adding new LLM models to Letta Code. Use when the user wants to add support for a new model, needs to know valid model handles, or wants to update model-specific compatibility behavior. Covers runtime catalog sources, CI test matrices, and handle validation.

Fetch PR review comments, verify each against real code/docs, fix valid issues, commit and push

Use when a research-writing task spans multiple sections, medium-sized revisions, full-paper drafting, or repeated quality failures

Use when writing literature review sections - guides searching, organizing, and synthesizing academic sources

Review documentation for quality, clarity, SEO, and technical correctness. Optimized for Docus/Nuxt Content but works with any Markdown documentation. Use when asked to: "review docs", "check documentation", "audit docs", "validate documentation", "improve docs quality", "analyze documentation", "check my docs", "review my documentation pages", "validate MDC syntax", "check for SEO issues", "analyze doc structure". Provides actionable recommendations categorized by priority (Critical, Important, Nice-to-have).

Maximum-saturation research orchestration: ALWAYS proposes the final materials first (PDF+DOCX default), then parallel explore+librarian swarms across codebase, web, official docs, and OSS repos — max-roster teammode when the harness has it — with live journaling, a recursive EXPAND loop driven by leads workers return in message text, empirical verification by running code, and a cited synthesis with charts/Mermaid/assets behind a mandatory visual-QA gate. ACTIVATES ONLY on an explicit user demand for research — the word 'ulw-research' ('/ulw-research', '$ulw-research'), any 'ulw' research wording, 'ultradebate' or 'hyperdebate' research requests, or an explicit request for research / deep research / an ultra-precise investigation, in any language. Never self-activates for ordinary questions, debugging, or implementation context-gathering. While active it overrides exploration-bounding defaults: exhaustive coverage is the goal.

Product Owner (Pax). Use for backlog management, story refinement, acceptance criteria, sprint planning, and prioritization decisions Triggers: validate story, backlog, acceptance criteria, close story, po, @po, prioritize, story draft. Use when the user runs /aiox-po or @po.

Test Architect & Quality Advisor (Quinn). Use for comprehensive test architecture review, quality gate decisions, and code improvement. Provides thorough analysis including requ...

GitHub Repository Manager & DevOps Specialist (Gage). Use for repository operations, version management, CI/CD, quality gates, and GitHub push operations. ONLY agent authorized...

Database Architect & Operations Engineer (Dara). Use for database design, schema architecture, Supabase configuration, RLS policies, migrations, query optimization, data modelin...

Business Analyst (Atlas). Use for market research, competitive analysis, user research, brainstorming session facilitation, structured ideation workshops, feasibility studies, i...

Create or redesign SaaS and AI product landing pages where a real workflow, interface, or deterministic demo is the central proof. Use for pale atmospheric shells, product UI in the hero, prompt-to-output stories, audience tabs, compact feature modules, honest pricing comparisons, and FAQ handoffs.

Create or redesign creative agency, production studio, architecture, culture, and portfolio websites with billboard typography, hard black-and-white chapters, exposed grids, documentary imagery, irregular collages, restrained parallax, brutalist navigation, and accessible FAQ controls.

Implement or tune Three.js game cameras. Use for isometric framing, follow behavior, orbit/zoom limits, occlusion, lock-on, camera shake, touch camera controls, and camera regression tests.

Use when the user asks Codex to research, find learning materials, process "素材：" links, "请你读" / "精读" a material, build a material radar, or use SenSight-like broad information retrieval for career learning and Agent infra tracking. Do not use the career/Agent-infra routing bias for user-directed `整理笔记` into a named note.

Comprehensive guide for AntV L7 geospatial visualization library. Use when users need to: (1) Create interactive maps with WebGL rendering (2) Visualize geographic data (points, lines, polygons, heatmaps) (3) Build location-based data dashboards (4) Add map layers, interactions, or animations (5) Process and display GeoJSON, CSV, or other spatial data (6) Integrate maps with AMap (GaodeMap), Mapbox, Maplibre, or standalone L7 Map (7) Optimize performance for large-scale geographic datasets

IP address utilities — geolocation, subnet scanning, port discovery. Zero barrier — use with x-cmd or curl + standard networking tools. Use for "ip", "geolocation", "subnet", "port scan", "network".

Task delegation to the heartbeat agent via HEARTBEAT.md. Use when scheduling follow-ups, recurring checks, or recording background jobs for later monitoring.

Write robust JavaScript for GitHub Actions github-script steps.

Instructions for downloading and installing .NET preview runtime versions required by the WinForms repository. Use when test executables fail with "framework not found" errors or when a specific .NET preview runtime version needs to be installed.