SD: Fill Skill Module — Autonomous Agents Agentic Workflow

schema_version: 1
id: agent_sd_coverage
title: "SD: Fill Skill Module"
description: "Find the lowest-coverage module, explore it in the browser, improve its skill files"
overview: |
  The Skill Discovery agent runs this chain to grow a pack's knowledge coverage over time. It scans the installed modules, picks the one with the weakest knowledge file — lowest confidence, outdated last-verified date, or known gaps — and then actively explores that area of the product in a live browser to update the knowledge file.

  The output is a PR on the skill-packs repo that raises confidence, adds missing selectors or states, and updates the verification timestamp. Dispatch it on a schedule — typically once a day — to keep long-lived packs accurate without manual curation.

category:
  level: pipeline
  domain: engineering

metadata:
  agent: true
  role: sd

params:
  hint:
    type: string
    default: ""
    description: "Optional extra instructions for the agent"
  entry_path:
    type: string
    default: "~/workspace"
    description: "Working directory for the agent"

steps:
  - type: terminal.open
    title: "Agent: SD Coverage"
    cwd: "{{entry_path}}"
  - type: terminal.run
    cmd: >-
      source ~/.agent-env && claude --dangerously-skip-permissions
      "learn about the target app and Skill Discovery methodology.
      review the skill pack coverage summary to find the module with lowest coverage or quality.
      navigate to that module on the target app, snapshot all states and interactions.
      update the module skill files, install with sidebutton registry update.
      commit, push, and create a PR to the skill-packs repo.
      {{hint}}
      report which module was improved and what was added."