wtfMS

Plugin wtfMS from marketplace wtf-ms in iowarp/wtf-ms. Commands read from the repo's manifest — 1 month ago.

Skills (6)

• wtfms-lab-definer

  Structure resource interviews into a feasibility map with gaps and alternatives.

• wtfms-literature-reviewer

  Review supplied materials and provided URLs with citation provenance and explicit gaps.

• wtfms-research-explorer

  Synthesize candidate materials-science questions and write researcher-confirmed identity.

• wtfms-task-executor

  Produce type-specific artifacts for one approved materials-science task and verify the files.

• wtfms-task-verifier

  Independently inspect research task outputs against their approved assumptions, artifacts, and evidence.

• wtfms-workflow-planner

  Build a resource-aware dependency plan with researcher assumptions and tagged defaults.

Slash commands (18)

• /wtfms:add-task

  Add a new task to the research workflow

• /wtfms:archive-task

  Archive a task — moves it to the Archived section of WORKFLOW.md, out of active execution

• /wtfms:checkpoint

  Save, restore, or list research state snapshots

• /wtfms:cost

  Show token usage and estimated cost per research step, collected passively during normal operation

• /wtfms:define-research-tasks

  Build a research workflow — select from traditional templates or create from scratch, with assumption interviews per task

• /wtfms:define-virtual-lab

  Interview to map all available lab resources — equipment, HPC, software, collaborations — into a VIRTUAL-LAB.md that guides feasible task planning

• /wtfms:execute-task

  Execute a specific research task or the entire workflow

• /wtfms:help

  Overview of wtf-MS commands, workflow, and philosophy

• /wtfms:identify-research

  Start exploring a research subject — guided interview to select fields, define scope, and generate a research prompt

• /wtfms:literature-review

  Conduct literature review, identify research gaps, and define search keywords

• /wtfms:pause-research

  Pause the current research workflow — saves state and marks active task as paused

• /wtfms:progress

  Show research progress with statusline, recent work, and smart routing to next action

• /wtfms:remove-task

  Remove a task from the research workflow permanently

• /wtfms:resume-research

  Resume research from a paused state — shows current position and suggests next action

• /wtfms:settings

  View and edit wtf-MS project configuration

• /wtfms:status

  Show full research project dashboard — identity, workflow progress, data, and next action

• /wtfms:upload-data

  Register and index data files, papers, or datasets for use in research tasks

• /wtfms:wtfp

  Bridge to wtf-p — translate research state into a paper project handoff, with include/exclude control

Subagents (5)

• wtfms-lab-definer

  Structures a researcher's resource description into VIRTUAL-LAB.md with domain-aware gap analysis and a resource-to-task mapping. Never interviews the researcher; reports missing details for the orchestrator to ask. Returns VIRTUAL LAB DEFINED or CHECKPOINT REACHED.

• wtfms-literature-reviewer

  Conducts structured literature review for material science research from supplied papers first and the web second. Maps current knowledge, identifies gaps, proposes keyword refinements, and signals loop-back if the prompt needs revision. Never interviews the researcher. Returns LITERATURE REVIEW COM

• wtfms-research-explorer

  Synthesizes a researcher interview into candidate research prompts, then writes RESEARCH.md and STATE.md once the researcher has chosen. Never interviews the researcher directly. Returns CANDIDATES READY, RESEARCH IDENTIFIED, CHECKPOINT REACHED, or EXPLORATION INCONCLUSIVE.

• wtfms-task-executor

  Executes individual research tasks by type — generates protocols, scripts, analysis code, literature summaries, or theoretical derivations — and verifies every output is on disk before reporting. Never interviews the researcher. Returns TASK COMPLETE, CHECKPOINT REACHED, or TASK BLOCKED.

• wtfms-workflow-planner

  Builds a research workflow from a selected template, the researcher's customizations, and the assumption answers the orchestrator already collected. Checks feasibility against VIRTUAL-LAB.md and orders tasks by dependency. Creates WORKFLOW.md. Never interviews the researcher. Returns WORKFLOW COMPLE

Hooks (2)

• Stop

  python3 "${CLAUDE_PLUGIN_ROOT}/wtf-ms/scripts/record_metrics.py"

• SubagentStop

  python3 "${CLAUDE_PLUGIN_ROOT}/wtf-ms/scripts/record_metrics.py"

Similar plugins