From kadai
Use INSTEAD OF superpowers:brainstorming when working in any repository that contains a `.kadai/` directory. Runs the full brainstorming flow (explore → questions → approaches → design → user approval → spec → self-review) AND then plumbs the result into the kadai spine: auto-creates the epic if missing, creates a feature, attaches the spec to the feature as `spec.md`. The single visible artifact at the end is a feature in `.kadai/epics/<E>/features/<F>/` with `spec.md` attached, NOT a free-floating `docs/superpowers/specs/*.md`. If you load `superpowers:brainstorming` in a kadai repo, you'll write specs into the wrong place and lose the spec → plan → implementation provenance kadai is designed to capture.
How this skill is triggered — by the user, by Claude, or both
Slash command
/kadai:kadai-brainstormingThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill wraps `superpowers:brainstorming` with kadai spine plumbing. The conversational question flow, the visual companion offer, the design presentation, and the user-approval gates are all unchanged — the upstream skill's content carries through.
This skill wraps superpowers:brainstorming with kadai spine plumbing. The conversational question flow, the visual companion offer, the design presentation, and the user-approval gates are all unchanged — the upstream skill's content carries through.
At the spec-writing step (after the user approves the design, before you would write to docs/superpowers/specs/...), do this instead:
kadai.list_epics() (MCP).kadai.create_epic(title=<inferred from brainstorm>, phase='mvp'). The brainstorming flow's "what are we building" answer is usually the epic title.kadai.create_feature(epic_id=<id>, title=<derived from spec>, phase=<epic's phase>)./tmp/, then call kadai.attach_spec(feature_id=<id>, source_path=<tmp-path>). This moves the file into the feature's directory as spec.md.spec.md and the user is reworking:
.kadai/epics/<E>/features/<F>/spec.md to spec.<ISO-timestamp>.md.bak in the same directory. (Use the kadai.attach_spec MCP tool if it supports a replace parameter; otherwise do the backup yourself via shell cp then call the standard attach_spec which overwrites.)The self-review pass is unchanged: scan for placeholders, internal contradictions, scope creep, ambiguity.
Tell the user: "Spec written and attached to FEAT-XXX. Read it at .kadai/epics/<E>/features/<F>/spec.md. Let me know if you want changes before we run /kadai-writing-plans."
When the user approves, invoke kadai-writing-plans (NOT superpowers:writing-plans).
Run this check at the START of the skill:
test -d .kadai && echo "kadai repo" || echo "non-kadai repo"
If non-kadai → fall through to superpowers:brainstorming and warn the user once: "This is kadai-brainstorming but no .kadai/ was found — invoking upstream brainstorming with no spine plumbing."
docs/superpowers/specs/. That's the upstream skill's path. Kadai's spine is .kadai/epics/<E>/features/<F>/spec.md.superpowers:brainstorming (do NOT load both — this skill subsumes it)kadai-writing-planskadai skill (passive guidance + MCP tool reference)Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub fintanislost/kadai --plugin kadai