{"id":"a8e2d293-2910-475c-979b-5866cfabb524","slug":"nmalinowski-project-task-router","kind":"skill","name":"project-task-router","summary":"Project task skill router for copilot","owner":"nmalinowski","repoUrl":"https://github.com/nmalinowski/project-task-router","license":"MIT","licenseTier":"permissive","trust":"community","stars":0,"match":null,"fewStars":true,"format":"skill","createdAt":"2026-02-25T22:28:03.000Z","addedAt":"2026-10-22T02:24:58.949Z","logoUrl":"https://avatars.githubusercontent.com/u/4306842?u=95b95decede7838d70e51f83af4e0b26a9cde959&v=4","linkOnly":false,"collection":null,"skillPath":null,"categories":["productivity"],"tags":[],"runtime":"binary","local":true,"cloudCapable":false,"cloudCapability":"unknown","cloudReadyOnPlatform":false,"needsApiKey":false,"reviewed":false,"syncedAt":"2026-10-23T22:37:03.860Z","sample":false,"url":"https://conduits.dev/tools/nmalinowski-project-task-router","language":"Python","transports":["stdio"],"install":{},"requirements":{"env":[],"upstreamMayCharge":false},"tools":null,"readme":"<p align=\"center\">\n  <img src=\"docs/logo.svg\" alt=\"Project Task Router Logo\" width=\"120\" height=\"120\">\n</p>\n\n# Project Task Router\n\n[![Deploy GitHub Pages](https://github.com/nmalinowski/project-task-router/actions/workflows/deploy-pages.yml/badge.svg?branch=main)](https://github.com/nmalinowski/project-task-router/actions/workflows/deploy-pages.yml)\n\n[**Live Router Demo**](https://nmalinowski.github.io/project-task-router/)\n\nCross-platform task triage plugin for Codex, Claude Code, and Copilot CLI.\n\nIt routes incoming work requests to:\n\n- A top domain route\n- A recommended skill stack (`primary` + `secondary`)\n- A model profile (`deep_reasoning`, `balanced`, `fast_execution`)\n- Clarification mode when intent is ambiguous\n\n## Features\n\n- Deterministic weighted router script: `skills/project-task-router/scripts/route_task.py`\n- 8-domain routing taxonomy\n- Token/phrase/regex signal scoring with negative-signal penalties\n- Ambiguity handling with `alternatives`\n- Dynamic installed-skill discovery with alias fallback\n- Missing-capability detection (`missing_capabilities`)\n- Subagent orchestration pattern for review then remediation\n\n## Domain Taxonomy\n\n1. `market-discovery-validation`\n2. `product-strategy-positioning`\n3. `architecture-build-planning`\n4. `feature-implementation`\n5. `ui-ux-frontend`\n6. `security-compliance-reliability`\n7. `data-analytics-ai`\n8. `platform-ops-reliability`\n\n## Requirements\n\n- Python 3.9+\n- Optional installed skill ecosystem:\n  - `nmalinowski/groundwork`\n  - `nmalinowski/agents`\n  - `nmalinowski/ui-ux-pro-max-skill`\n  - `21st.dev/community/components`\n  - `nano-banana-pro`\n\n## CLI Usage\n\nBasic routing:\n\n```bash\npython skills/project-task-router/scripts/route_task.py --task \"Implement HIPAA-compliant audit logging\"\n```\n\nMarkdown output:\n\n```bash\npython skills/project-task-router/scripts/route_task.py --task \"Plan API architecture\" --format markdown\n```\n\nRoute with explicit installed-skill list:\n\n```bash\npython skills/project-task-router/scripts/route_task.py --task \"Improve frontend layout\" --installed-skills \"ui-ux-pro-max,design-system,product-design\"\n```\n\nRoute with installed-skill file:\n\n```bash\npython skills/project-task-router/scripts/route_task.py --task \"Build RAG pipeline\" --installed-skills-file installed-skills.json\n```\n\n## Skill Discovery Order\n\nWhen `--installed-skills` and `--installed-skills-file` are not provided, discovery runs in this order:\n\n1. Environment variables: `ROUTER_INSTALLED_SKILLS`, `INSTALLED_SKILLS`, `CODEX_INSTALLED_SKILLS`, `COPILOT_INSTALLED_SKILLS`, `CLAUDE_INSTALLED_SKILLS`\n2. Relative manifests:\n   - `installed-skills.json`\n   - `.codex/installed-skills.json`\n   - `.claude/installed-skills.json`\n   - `copilot/installed-skills.json`\n3. Home manifests:\n   - `~/.codex/installed-skills.json`\n   - `~/.claude/installed-skills.json`\n4. Plugin-local manifests:\n   - `<plugin-root>/installed-skills.json`\n   - `<plugin-root>/.codex/installed-skills.json`\n   - `<plugin-root>/.claude/installed-skills.json`\n   - `<script-dir>/installed-skills.json`\n\nFirst existing manifest wins.\n\n## Output Fields\n\n- `route`: selected domain or `needs-clarification`\n- `confidence`: `low`, `medium`, `high`\n- `primary_skills` and `secondary_skills`\n- `alternatives`: close competing routes\n- `missing_capabilities`: required capability IDs not resolved from known inventory\n- `evidence`: domain score map\n- `note`: routing/discovery hints\n\n## Subagent Workflow\n\nUse this sequence when executing routed work:\n\n1. Run router and confirm route/model/skills.\n2. Spawn review subagent for comprehensive findings (correctness, regressions, security, tests).\n3. After review completes, spawn remediation subagent.\n4. Remediation policy:\n   - Auto-fix `low`, `medium`, `high`\n   - Escalate `critical` findings for explicit approval\n\n## Development\n\nRun tests:\n\n```bash\npython -m unittest discover -s tests -v\n```\n\n## License\n\nMIT (see `plugin.json`).\n"}