skills/kanban/SKILL.md
Persistent task management through kanban boards with cross-session tracking. Provides guidance for using wicked-garden:kanban for CRUD operations, dependencies, priorities, and sprint management. Use when: "track tasks", "create a task", "add a todo", "manage tasks", "show my tasks", "what should I work on", "move task to done", "update task status", "kanban board", "project management"
npx skillsauth add mikeparcewski/wicked-garden task-managementInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Persistent task management through Python scripts and a web UI, providing visual kanban boards and cross-session task tracking.
Wicked Kanban provides AI-native task management that persists across Claude Code sessions. Use the kanban.py script via Bash for CRUD operations, and the web UI for visual management.
Key benefit: Tasks created with TodoWrite are automatically synced to the kanban board via PostToolUse hook, but using the kanban script directly provides richer features (dependencies, priorities, sprints).
The script is located at ${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py. Execute commands via Bash.
List all projects:
sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py" list-projects
Get full project state (swimlanes + tasks):
sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py" get-project PROJECT_ID
Create a project first:
sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py" create-project "Project Name" -d "Description"
Create a task (need swimlane_id from project):
sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py" create-task PROJECT_ID "Task Name" SWIMLANE_ID -p P1 -d "Description"
Priority levels: P0 (Critical), P1 (High), P2 (Normal), P3 (Low)
Update task status (move to different swimlane):
sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py" update-task PROJECT_ID TASK_ID --swimlane NEW_SWIMLANE_ID
Update task priority:
sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py" update-task PROJECT_ID TASK_ID --priority P0
Add comment to task:
sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py" add-comment PROJECT_ID TASK_ID "Comment text"
Link commit to task:
sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py" add-commit PROJECT_ID TASK_ID abc123
Search tasks across projects:
sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py" search "search query"
Search within a specific project:
sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py" search "search query" --project PROJECT_ID
Four board types with purpose-built column schemas:
| Board Type | Terminal Column | Best For |
|------------|----------------|----------|
| crew (default) | done (Done) | Software development, crew phases |
| jam | jam:decision_made (Decision Made) | Brainstorming, design decisions |
| collaboration | collab:complete (Complete) | Cross-team workflows |
| issues | done (Done) | Bug tracking, general issues |
Create a typed initiative with --board-type:
sh "${CLAUDE_PLUGIN_ROOT}/scripts/_python.sh" "${CLAUDE_PLUGIN_ROOT}/scripts/kanban/kanban.py" create-initiative PROJECT_ID "API Design" --board-type jam
# or: /wicked-garden:kanban:initiative create "API Design" --board-type jam
Filter board-status: /wicked-garden:kanban:board-status --board-type jam
Reaching a terminal column on jam or collaboration boards auto-writes a wicked-garden:mem record (decision or finding). Crew and issues boards do not trigger mem writes.
See refs/scoped-boards.md for full column schemas, provisioning rules, and CLI examples.
TodoWrite calls are automatically synced to a "Claude Tasks" project on the kanban board via the PostToolUse hook. Tasks are organized into session-based sprints for better organization.
Use /wicked-garden:kanban:name-session "Feature Name" to name the session sprint. Each session gets its own sprint for easy filtering.
Access the visual kanban board at http://localhost:18888. The web UI provides:
The server auto-starts with the Claude Code session. To start manually:
cd "${CLAUDE_PLUGIN_ROOT}" && uv run wicked-garden:kanban
See refs/workflow-patterns.md for complete workflow examples (feature dev, bug fix, session resume).
| Action | Command |
|--------|---------|
| List projects | list-projects |
| Get project | get-project PROJECT_ID |
| Create project | create-project "Name" -d "Desc" |
| Create task | create-task PROJECT_ID "Name" SWIMLANE_ID |
| Update task | update-task PROJECT_ID TASK_ID --swimlane ID |
| Add comment | add-comment PROJECT_ID TASK_ID "Text" |
| Link commit | add-commit PROJECT_ID TASK_ID HASH |
| Add artifact | add-artifact PROJECT_ID TASK_ID "Name" --type file |
| Project comment | add-project-comment PROJECT_ID "Text" |
| Search | search "query" |
| Create initiative | create-initiative PROJECT_ID "Name" --board-type jam |
| List initiatives | list-initiatives PROJECT_ID --board-type jam |
refs/api-reference.md — Core script commands (projects, tasks, initiatives, search)refs/api-advanced.md — Advanced features (sprints, artifacts, data model)refs/scoped-boards.md — Full column schemas, provisioning, wicked-garden:mem triggersrefs/workflow-patterns.md — Common workflow examples (feature dev, bug fix, session resume)refs/advanced-patterns.md — Sprint planning, dependency graphs, priority filteringexamples/feature-workflow.md — End-to-end feature development workflow exampledevelopment
--- name: large-scale-migration description: How to execute a LARGE MECHANICAL change across any codebase with LEVERAGE instead of an agent-grind or hand-edits — a cross-cutting migration, refactor, rename, dialect/framework/DB port, library adoption, or bulk transform. The map→transform→gate pattern: a deterministic transform driven by a source-of-truth map, proven by a differential-equivalence gate. Use when the work is "migrate all X to Y", "rename Z everywhere", "port to a new DB/dialect/fra
testing
v11 LLM-based work-shape classifier. Replaces the regex archetype detector with the model's own reasoning. Reads the user's prompt, picks the right archetype(s) from the catalog, identifies signals (blast_radius, novelty, reversibility, etc.), and persists to SessionState so subsequent turns steer correctly. Use when: the prompt_submit hook emitted a `<wg classify-due />` directive, OR explicitly invoked at session start, OR when re-classifying after the user changes scope mid-session.
tools
v11 work-shape archetype runner. When a prompt has been routed to one of the 9 archetypes (triage, explore, specify, decide, ship, review, incident, build, migrate), this skill is the entry point. It picks the right per-archetype playbook from refs/ and executes the phase shape declared in `.claude-plugin/archetypes.json`. Use when: a `<wg archetype="X">` or `<wg archetypes>` system-reminder tag appears, an explicit "let's run the X archetype" request, or when one of the per-archetype slash commands resolves to this skill.
development
Show or set the session intent variable. Intent gates how loud the framework is — simple-edit (silent), feature/research (synthesis directive), rigor (full crew context). Auto-detected on turn 1; this skill overrides explicitly. Sticky for the session. Use when: "set intent", "intent override", "/wicked-garden:intent", "make the framework quiet", "force rigor", "what's my intent".