Update to Claus Argos Skill OS

{
  "name": "capture-workflow-as-skill",
  "description": "Observe, extract, normalize, and convert a repeated real-world workflow into a reusable agent skill with triggers, inputs, decisions, tools, permissions, resources, quality gates, exceptions, and evaluation cases. Use when a user repeatedly explains the same task, asks to save a workflow as a skill, convert a conversation or SOP into a skill, learn from completed work, standardize a successful process, or improve a skill from usage history with explicit human review.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 245
    },
    {
      "relative_path": "assets/workflow-intake.md",
      "size_in_bytes": 472
    },
    {
      "relative_path": "references/workflow-extraction.md",
      "size_in_bytes": 1123
    }
  ],
  "skill_md_contents": "---\nname: capture-workflow-as-skill\ndescription: Observe, extract, normalize, and convert a repeated real-world workflow into a reusable agent skill with triggers, inputs, decisions, tools, permissions, resources, quality gates, exceptions, and evaluation cases. Use when a user repeatedly explains the same task, asks to save a workflow as a skill, convert a conversation or SOP into a skill, learn from completed work, standardize a successful process, or improve a skill from usage history with explicit human review.\n---\n\n# Capture Workflow as Skill\n\nCapture demonstrated work without silently learning unrelated behavior.\n\n## Workflow\n\n1. Confirm the workflow boundary, intended users, frequency, variability, risk, source material, and installation location.\n2. Gather representative successful and failed examples. Do not treat a single run as a stable process.\n3. Extract objective, trigger language, inputs, outputs, invariants, decisions, tools, permissions, external effects, failure modes, approvals, and completion evidence.\n4. Separate organization-specific facts from general method and secrets from reusable configuration.\n5. Decide whether the durable artifact should be a skill, prompt, SOP, agent, automation, hook, connector, or combination. Do not force every repetition into a skill.\n6. Design the smallest reliable skill using `references/workflow-extraction.md` and `assets/workflow-intake.md`.\n7. Reuse scripts, templates, and references when deterministic execution or progressive disclosure improves reliability.\n8. Create realistic should-trigger, should-not-trigger, execution, missing-tool, and safety tests.\n9. Obtain explicit approval before installing, activating, replacing, or granting new permissions.\n10. Forward-test, repair demonstrated failures, version the result, and record provenance.\n\n## Learning boundary\n\n- Never retain private material or preferences beyond the authorized artifact.\n- Never infer permission from observed user behavior.\n- Never self-update an installed skill without a reviewable diff and approval.\n- Exclude accidental workarounds, obsolete steps, credentials, and one-off exceptions unless intentionally generalized.\n\n## Output\n\nReturn workflow model, artifact decision, proposed skill architecture, risks, evaluation set, approval points, and final package when creation is authorized.\n\nFor workflows that create downstream controlled artifacts, define an artifact contract: stable identity, version, lifecycle status, owner, sources, usage, precedence, dependencies, acceptance evidence, and replacement history. The captured workflow must be executable without hidden conversation context and include a clean-room test plus observable completion evidence.\n"
}