一键导入
todo
TODO tracking system for spec TODOs. Use when creating, updating, or querying spec TODOs in docs/spec/todos/.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
TODO tracking system for spec TODOs. Use when creating, updating, or querying spec TODOs in docs/spec/todos/.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
Use `mdvs search` for any content lookup in a markdown directory — semantic / hybrid / SQL-filtered, beats Grep / Glob for finding by meaning. `mdvs init` infers a schema from existing markdown, `mdvs check` validates frontmatter, `mdvs update` evolves the schema as the KB grows. Activate whenever the project contains markdown with frontmatter, whether or not `mdvs.toml` exists yet.
Release process for mdvs. Use when the user asks to make a release (patch, minor, major, or release candidate).
Use when the user asks to commit, commit and push, or make a git commit. Covers branching, commit workflow, conventional commits, TODO updates, and push/PR rules.
Use when writing, modifying, or removing Rust code in the mdvs codebase. Covers implementation workflow, testing, verification, and downstream updates (specs, example_kb, mdbook).
mdBook documentation conventions. Apply when writing, editing, or reviewing pages in book/src/. Covers content rules, example verification, tone, and structure.
GitHub Issues + Project tracking. Use when creating, updating, or querying issues and managing the project board.
| name | todo |
| description | TODO tracking system for spec TODOs. Use when creating, updating, or querying spec TODOs in docs/spec/todos/. |
Spec TODOs live in docs/spec/todos/ as individual markdown files with YAML frontmatter.
docs/spec/todos/index.md — Single table sorted by ID with status, priority, and datesdocs/spec/todos/TODO-NNNN.md (zero-padded to 4 digits)glob docs/spec/todos/TODO-*.md---
id: <next_id>
title: <short title>
status: todo # todo | in-progress | done
priority: high # high | medium | low
created: <YYYY-MM-DD>
depends_on: [] # list of TODO IDs that must complete first
blocks: [] # list of TODO IDs this blocks
---
# TODO-NNNN: <Title>
## Summary
<1-3 sentences describing the goal>
## Details
<Full description: steps, design questions, files to create/update>
index.md: Add row to the table (sorted by ID)todo → in-progress)status: in-progressin-progress → done)status: done
completed: <YYYY-MM-DD>
files_created: [...] # files created during resolution
files_updated: [...] # files modified during resolution
## Details with ## Resolution containing the resolution notesWhen a TODO is fully covered by another TODO's implementation:
status: done
completed: <date of subsuming TODO's completion>
subsumed_by: <subsuming TODO ID>
## Details to ## Original Scope, add:
## Resolution
Subsumed by [TODO-NNNN](TODO-NNNN.md). <brief explanation>.
subsumed:
- <absorbed TODO ID>
depends_on: [8] means "cannot start until #8 is done"blocks: [9] means "#9 cannot start until this is done"Only populated on completion:
files_created: New files created as part of the resolutionfiles_updated: Existing files modified as part of the resolutioncommands/check.md, workflows/inference.md)TODO files live at docs/spec/todos/. Adjust relative links accordingly:
../terminology.md, ../storage.md, ../commands/check.md, ../workflows/inference.mdTODO-NNNN.md (same directory)