{"id":17576,"plugin_id":"plugins_6a78e83987748191afc0c56e12172fce","kind":"skill","collection_source":null,"comparison_source":null,"observed_at":"2026-09-30T23:14:16.179Z","digest":"0f02f2d24c901cd5a9b8bf034c30499444e7f8f76b851c2f1eb1d188d8b4b49c","against":null,"payload":{"description":"Specify recurring operational, review, or content workflows with explicit triggers, inputs, actions, and ownership. Use when designing repeatable team routines, approval loops, release checklists, or automated processes — even if the user says \"design a workflow for this\". Do NOT use for one-off coding tasks.","included_files":[{"relative_path":"agents/openai.yaml","size_in_bytes":101}],"name":"workflow-designer","skill_md_contents":"---\nname: workflow-designer\ndescription: \"Specify recurring operational, review, or content workflows with explicit triggers, inputs, actions, and ownership. Use when designing repeatable team routines, approval loops, release checklists, or automated processes — even if the user says \\\"design a workflow for this\\\". Do NOT use for one-off coding tasks.\"\n---\n\n# Workflow Designer\n\nDesign and formalize repeatable human-in-the-loop and autonomous workflows by identifying cyclical operational loops, pushing checkpoints right, and defining unambiguous execution contracts.\n\n---\n\n## Core Invariants\n\n1. **Push Checkpoints Right**: Defer human checkpoints as far downstream as possible; maximize autonomous work before requesting review.\n2. **Actionable Decision Briefs**: Checkpoints must present a tight, decision-ready brief with clear diffs and links—never raw, uncurated outputs.\n3. **Mandate Nothing Structural**: Only add AI agents, schedules, or approval gates when the problem domain strictly requires them.\n4. **Self-Contained Spec Completeness**: A workflow specification is complete only when an implementer agent can build it without asking clarifying questions.\n5. **Durable Notes Separation**: Distinguish between immutable workflow specifications (`workflows/*.md`) and evolving domain observations (`NOTES.md`).\n\n---\n\n## Architecture & Map of Content (MOC)\n\n```\n[ Operational Habit / Team Routine ] ──► [ Loop Identification & Grilling ] ──► [ Workflow Formalization ] ──► [ Implementer Handoff ]\n```\n\n| Component | Responsibility | Location |\n|---|---|---|\n| **Workflow Specification** | Executable contract (trigger, inputs, actions, checkpoints) | `workflows/*.md` |\n| **Domain & Tools Notes** | Observed vocabulary, habits, tools, and channels | `NOTES.md` |\n| **Grilling Protocol** | Socratic distillation of requirements and boundary edges | `skills/grilling/SKILL.md` |\n\n---\n\n## Step-by-Step Procedure (TWI)\n\n### Step 1: Identify Cyclical Loops & Document Baseline Notes\n- **Action**: Analyze the user's recurring tasks (daily, weekly, per-release) and record observed tools/channels into `NOTES.md`.\n- **Key Point**: Map the natural lifecycle from trigger to final delivery.\n- **Why**: Capturing the current baseline prevents designing theoretical workflows that mismatch daily reality.\n\n### Step 2: Conduct Socratic Grilling on Workflow Boundaries\n- **Action**: Run a focused grilling round exploring:\n  - **Trigger**: Event-based (e.g. webhooks, new issue) vs. time-scheduled (e.g. cron).\n  - **Inputs & State**: Exactly what artifacts must be ingested.\n  - **Actions**: Discrete operational or agentic transformations.\n  - **Checkpoints**: Minimal review gates pushed to the end.\n- **Inline Checklist**:\n  - [ ] Triggers clearly specified with fallback polling intervals\n  - [ ] Checkpoints provide decision-ready briefs\n  - [ ] Failure paths and escalation owners defined\n\n### Step 3: Author Executable Workflow Spec\n- **Action**: Write the workflow contract to `workflows/<name>.md` detailing trigger, inputs, actions, checkpoint briefs, and outputs.\n- **Key Point**: Ensure an implementer agent could implement scripts or automations from the document alone.\n- **Why**: Ambiguity in workflow contracts causes brittle automation failures.\n\n---\n\n## Anti-Rationalization Guardrails\n\n| Tempting Rationalization | Binding Rule | Engineering Rationale |\n|---|---|---|\n| *\"Ask the human to verify every intermediate step.\"* | **Push checkpoints right; batch human reviews.** | Frequent interruptions cause human fatigue and destroy automation efficiency. |\n| *\"Dump raw log files for the human checkpoint.\"* | **Produce synthesized, decision-ready briefs.** | Humans review clean executive summaries 10x faster than raw debug logs. |\n| *\"Assume an LLM agent is required for every workflow step.\"* | **Use deterministic scripts where possible; AI only where judgment is needed.** | Deterministic scripts are faster, cheaper, and more reliable than LLMs for structured operations. |\n\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}