From pds
Guides PDS contributions: whitepaper alignment checklist, cross-reference updates, ADRs, and shipping via /pds:finish. Use before modifying skills, agents, or workflows.
How this skill is triggered — by the user, by Claude, or both
Slash command
/pds:contributeThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Process for modifying PDS itself. PDS is a living system — skills, agents, and the whitepaper co-evolve. Changes to one often require updates to others.
Process for modifying PDS itself. PDS is a living system — skills, agents, and the whitepaper co-evolve. Changes to one often require updates to others.
Read docs/whitepaper.md before changing workflow. The whitepaper is the source of truth for the Agentic SDLC model. Your change must align with it — or update it.
Implement the feature, fix, or improvement.
If your change affects any of these, update docs/whitepaper.md:
Not every PDS change needs a whitepaper update. Skip it for:
When adding a skill or agent, update all reference points:
CLAUDE.md — skills tableREADME.md — skills table/pds:team — if agent changes/pds:swarm — if workflow phase changesdocs/skills.md — skills catalogskills/*/EVAL.md — if modifying a skill with an eval, update scenarios if behavior changeddocs/whitepaper.md — if new terms introducedFor significant design decisions, store ADRs in docs/adr/. Filename: NNNN-brief-description.md. Format: Context, Decision, Consequences. Write when choosing between multiple valid approaches or making hard-to-reverse decisions.
Run /pds:finish to finalize — verifies, bumps version, commits, and pushes:
/pds:finish — Branch completion and shipping/pds:bump — Version bump and changelog (no push)npx claudepluginhub rmzi/portable-dev-system --plugin pdsApplies targeted improvements to an existing pm-skills skill based on feedback, validation reports, or convention changes. Reads current files, previews proposed changes, writes on confirmation, and suggests a version bump.
Guides project documentation structure, content requirements, and best practices. Includes templates for README, ARCHITECTURE, API docs, and change history.
Crafts effective prompts, CLAUDE.md instructions, rules, agent system prompts, and skill bodies for reliable Claude Code behavior.