dami

Plugin dami from marketplace dami-stack in dev-dami/dami-stack. Commands read from the repo's manifest — 3 days ago.

Skills (22)

• article-workshop

  Use when the user starts a piece of writing or has raw material to shape into an article. Not for code docs or agent docs.

• purposeful-tutor

  Use when the user wants to learn a topic over time rather than have it done. Not for one-off explanations.

• baton-pass

  Use when a chunk of work ends, the session runs short, work moves to another agent, tool or person, or you receive a packet. Skip mid-chunk. Starting an unattended agent needs OK.

• crew-lead

  Use when a ready chart, several cards or a long-running goal should go to a team of agents. Merging, opening PRs and outward steps need OK.

• agent-prose

  Use when creating or editing anything an agent reads: a skill, instruction file, packet template or agent doc. Not for docs written for people.

• house-rules

  Use when you need the stack's shared rules: gates, evidence labels, artifact shapes, team levels or routes, or the user is unsure how to proceed or asks what exists.

• module-depth

  Use when creating a module, reshaping an interface, placing a seam, or code keeps resisting change or tests. Not for feature work that fits the existing shape.

• shared-words

  Use when terms clash or stay vague, a costly-to-reverse decision lands, or you edit the glossary or decision records. Reading them doesn't count. New docs in a repo: offer first.

• failing-test-first

  Use when adding behaviour or fixing a bug where a test runner exists, even if tests went unmentioned. Skip throwaway work, docs, config and anything the user excluded.

• root-cause-hunt

  Use when a first fix failed, the cause isn't obvious, a failure is intermittent, or something got slower. Skip obvious one-liners.

• finish-line

  Use before calling multi-part or risky work done, committing a multi-file change, or opening a PR, and when reviewing someone's work. Not mid-task or for one-line edits.

• merge-case

  Use when writing or updating a pull request body, once the work has passed finish-line. Publishing it needs OK.

• intake-desk

  Use when working through issues or pull requests the user didn't author. Not for cards from work-chart. Posting, labelling, closing or running external code needs OK.

• question-tree

  Use when a plan hides choices that change the result with no sensible default, the user wants their thinking tested, or a decision waits on another person. Skip what defaults can handle.

• source-trail

  Use when work depends on docs, API or library facts the code can't confirm and the answer is worth keeping. Not for facts the repo itself settles.

• throwaway-spike

  Use when a screen's look or a state model's soundness is unclear and disposable running code would settle it sooner than talk. Skip settled designs. Side-branch commit needs OK.

• work-chart

  Use when agreement needs a durable plan, work needs slicing into a graph, or a huge effort has key decisions open. Skip one-session work. Publishing needs OK.

• after-action

  Use after repeated corrections or avoidable failures, after a hard bug, or before clearing a long session. Proposes only. Not after a clean run.

• automation-blueprint

  Use when repeated work should be delegated and its spec isn't clear, or when designing agents that turn tickets into merged code. Design only, no build.

• host-connect

  Use when the user wants work to run on their VPS or another machine, or a host is missing or broken in team_host. Walks them through connecting it.

• repo-fitting

  Use when a tracker step finds no repo config, or a repo lacks a guard it keeps needing: commit gate, import rules, git safety. Installing or editing config needs OK.

• walk-me-through

  Use when a step truly needs a human: credentials, dashboards, provisioning, one-off cutovers. Never for what an agent can do. Show the script before anyone runs it.

Slash commands (15)

• /dami:build

  Build one card of agreed work

• /dami:chart

  Turn agreement into a plan and slice it into cards

• /dami:check

  Verify the current work against what was asked

• /dami:connect

  Connect a VPS or other machine so agents can run there

• /dami:crew

  Run a chart with a team of agents

• /dami:debrief

  Review this session for lasting fixes

• /dami:dig

  Investigate a question with every claim cited

• /dami:fix

  Find the confirmed cause of a hard bug and fix it

• /dami:goal

  Agree and pursue a long-running goal

• /dami:help

  What exists, and what fits now

• /dami:learn

  Learn a topic over time with a tutor

• /dami:pass

  Hand work on to a session, agent, tool or person

• /dami:settle

  Settle open decisions in a plan

• /dami:ship

  Check the work, then draft the pull request

• /dami:spike

  Answer a design question with throwaway code

MCP servers (1)

• dami

  runs: node ${CLAUDE_PLUGIN_ROOT}/bin/dami-mcp.js

Hooks (1)

• SessionStart

  node "${CLAUDE_PLUGIN_ROOT}/bin/session-context.js"