id: task-302 title: Support flexible ID formats for tasks and docs status: Done assignee:

• '@codex' created_date: '2025-10-17 22:09' updated_date: '2025-10-18 20:30' labels: [] dependencies: []

Description

Align ID parsing with Issue #404 requirements so CLI, MCP, and APIs accept variations. Implement parsing normalization once in shared utilities and ensure both task and document lookups use it.

Acceptance Criteria

• [x] #1 Task lookup accepts TASK-, task-, bare numeric id, and zero-padded variants.
• [x] #2 Document lookup accepts DOC-, doc-, bare numeric id, and zero-padded variants.
• [x] #3 Tests cover new parsing helper for both tasks and documents.

Implementation Plan

1. Add shared ID normalization helpers that handle prefix casing and numeric padding.
2. Refactor task/document lookup paths (CLI, core, filesystem, MCP, server) to use the helpers.
3. Expand unit and integration tests to cover uppercase/padded inputs across tasks and documents.

Implementation Notes

• Task and document comparisons now rely on dedicated helpers (src/utils/task-path.ts and src/utils/document-id.ts) so every caller works through a single normalization/equality path.

• CLI commands use Core.getDocumentContent/Core.loadTaskById to avoid touching the filesystem directly; loadTaskById exists specifically to bypass long-lived watchers so short-lived CLI processes (and Windows CI) exit cleanly.

• ID normalization is applied at construction-time (task creation, document saves) and whenever dependencies/parents are parsed, preventing accidental mixed-prefix storage.

• Added MCP/server bindings to those helpers, so task/document tools accept case-insensitive and zero-padded IDs without duplicating logic.

• Pending: coordinate with Claude for an additional review as requested.

• Verified flexible ID handling for CLI, MCP, server, and filesystem pathways.