planning-doc-triad
Planning Doc Triad
Use three planning documents with non-overlapping roles. If the repo already has canonical files for these roles, reuse them. If it does not, default to:
docs/roadmap.md: strategic direction and phased sequencingTODO.md: near-term actionable queuePLAN.md: one active detailed plan for the current job
This structure separates strategy, near-term execution, and active working context. Treat it as a hierarchy:
- roadmap sets direction
- todo holds the near-term candidate queue
- plan expands the one todo item currently being executed
When To Apply It
Apply this skill when:
- roadmap items are mixed with implementation tasks
- there is no clear "what next" list
- multiple plan files keep going stale
- detailed planning is happening only in chat history
- priorities are hard to infer from the current docs
Use it both for new repos and for cleaning up an existing planning system.
Core Rules
Roadmap doc
Use the roadmap for:
- product or technical direction
- phases or milestones
- intentional deferrals
- sequencing across multiple chunks of work
Do not use the roadmap for:
- session-level checklists
- fine-grained task breakdowns
- ephemeral notes
If the repo already has a canonical roadmap file, reuse it.
Otherwise default to docs/roadmap.md.
Todo doc
Use the todo file for:
- near-term candidate work
- active or next-up work
- unrelated items that could reasonably be worked on soon
- small actionable items
- ordered execution priority
Keep it:
- short
- concrete
- easy to scan
Do not turn it into a strategy memo or a storage place for distant ideas. Do not expand a todo item into a detailed execution document until it becomes the active job.
If the repo already has a canonical todo file, reuse it.
Otherwise default to TODO.md.
Active plan doc
Use the plan file for:
- one active initiative
- the one todo item currently being executed
- one detailed job breakdown
- current scope, non-goals, and steps
Keep one clearly active detailed plan file.
If the repo already has a canonical current-plan location, reuse it instead of adding another root-level plan file.
When a todo item becomes the current job, expand it here instead of turning TODO.md into prose.
Rewrite the active plan freely as the current job changes.
Do not preserve stale plan history in multiple competing active plan files.
Setup Workflow
When introducing this methodology to a repo:
- Check whether the repo already has a roadmap, todo list, or plan docs.
- Reuse existing files when they already match the intended role.
- If the repo has no established convention for a missing role, create the missing files in these default locations:
docs/roadmap.mdTODO.mdPLAN.md
- Remove overlap so each file has one clear job.
- Move durable decisions into canonical docs such as architecture or product docs.
Maintenance Workflow
When updating planning docs during normal work:
- Keep the roadmap file focused on strategic direction, milestones, and sequencing.
- Keep the todo file as the ordered near-term queue derived from that direction.
- When one todo item becomes the current job and needs detail, expand it into the active plan file.
- During execution, update the active plan as scope or steps change.
- After execution, remove completed items from the todo file, add any new follow-up tasks there, and rewrite the active plan for the next job.
- If the plan produced durable decisions, persist them outside the active plan file.
Writing Standards
- Keep all three files concise.
- Lead with what matters now.
- Use direct, concrete wording.
- Prefer ordering over categorization unless categories materially help.
- Avoid repeating the same item across files.
Recommended File Shapes
docs/roadmap.md
Prefer this structure:
- purpose or current recommendation
- phases or milestones
- likely deliverables per phase
TODO.md
Prefer this structure:
- one short rule line
## Next Up- a flat ordered list of near-term candidate tasks
PLAN.md
Prefer this structure:
- current job promoted from
TODO.md - goal
- in scope
- out of scope
- work plan
- exit criteria
Anti-Patterns
Avoid these failure modes:
- turning the roadmap into a backlog
- turning the todo file into a strategy memo
- keeping several plan files for parallel jobs without a clear active one
- leaving the active plan file as stale history after the job changes
- storing important product decisions only in the active plan file
Output
When applying this methodology, produce:
- the updated files, or the new files that were actually needed
- a short explanation of what each file now owns
- the current next actions
More from sjunepark/custom-skills
summarize
Use the steipete/summarize CLI to summarize URLs, local files, stdin, YouTube links, podcasts, and media with LLM models. Trigger when users ask to install or run summarize, configure model/provider API keys, tune output flags (length/language/json/extract/slides), set defaults in ~/.summarize/config.json, or troubleshoot summarize CLI errors.
41skills-cli
Operate the skills CLI to discover, install, list, update, remove, and initialize skills for Codex, Claude Code, and Pi. Use when users ask to manage skills from skills.sh, restore from lock files, sync skills from node_modules, or troubleshoot agent/installation scope (project vs global).
37post-implementation-review
Manually review already-implemented code for design flaws, abstraction issues, structural problems, or refactors that only became clear in real code. Use only when the user explicitly asks for a post-implementation review, explicitly asks whether recent implementation work revealed design or structure problems, or explicitly wants refactor recommendations after the code exists. Do not auto-trigger for ordinary implementation, debugging, explanation, or generic code review requests. Prefer embedded snippets with file-path comments over editor-oriented file and line references. Treat findings as signals about code shape and quality; prioritize root-cause design, ownership, abstraction, and organization improvements, including broad refactors when warranted, over bandage fixes such as tiny helper extractions or local polish.
30architecture-md-writer
Create, update, review, and split ARCHITECTURE.md files that explain a codebase's shape, major components, runtime flow, code map, and important invariants. Use when a repository lacks architecture docs, an existing ARCHITECTURE.md is stale or too detailed, a subsystem needs its own nested ARCHITECTURE.md, or a root architecture doc should link to deeper module architecture docs.
27agents-md-writer
Create, edit, review, and improve AGENTS.md files for repositories used by agentic coding tools with concise, actionable instructions and correct precedence behavior. Use whenever AGENTS.md content is being changed, including updating existing guidance, drafting a new AGENTS.md, migrating legacy instruction files, defining nested overrides in monorepos, or debugging why tools load the wrong guidance.
26source-investigator
Investigate external libraries, frameworks, and unfamiliar repositories by cloning the exact repo into a project-local temp workspace, ignoring that workspace in git, and delegating code reading to focused subagents so the main thread stays clean. Use whenever docs are incomplete, version-specific behavior matters, you need to learn how a codebase works, or exploring lots of source inline would pollute the main context.
24