
from task-orchestrator177
Turn conversation context into MCP work items: infer type/priority, anchor into containers (Bugs, Features, Tech Debt), create single items or work trees, and p
This skill converts natural-language conversation context into structured MCP work items. It infers a title, type (bug/feature/tech-debt/observation/action), priority, and scope, then scans existing containers to anchor the new item under the correct category. When appropriate it can create a work tree (root + subtasks) and pre-fill required notes using guidance from expectedNotes responses. The skill includes logic to read local config for schema keys and traits to enforce proper item typing.
Use this skill when the conversation surfaces actionable work: bug reports, feature ideas, tech-debt notes, follow-ups, or when the user explicitly asks to "track this", "log this bug", "create a task for", or "add this to the backlog". It is appropriate for agents that manage persistent task tracking and need to map freeform requests into a team's container structure.
Designed for agents integrated with MCP-style task APIs (Claude, Claude Code, Cursor) or any assistant that can call manage_items/manage_notes endpoints.
Task orchestrator skill for creating MCP work items from conversation context. No bundled scripts — pure instruction skill. SKILL.md is well-written with clear decision trees for container anchoring, schema discovery, and type inference. Minor security concern around reading file paths from env vars without validation. Niche usefulness as it requires the task-orchestrator MCP server.
Clean instruction-only skill. No scripts to test. Well-structured with good error handling guidance. Category mapping tables are practical and clear.