when-automating-workflows-use-hooks-automation

STANDARD OPERATING PROCEDURE

Purpose

Design and maintain hook-based automations (pre/post hooks, webhooks) with explicit constraints, monitoring, and confidence-aware delivery.

Trigger Conditions

• Positive: creating/updating hooks, chaining systems via callbacks, adding validation to hook payloads, retry/backoff tuning, telemetry for hook events.
• Negative: manual single runs, prompt-only edits (route to prompt-architect), or new skill weaving (route to skill-forge).

Guardrails

• Skill-Forge structure-first: keep SKILL.md, examples/, tests/ in place; add resources//references/ or log remediation tasks.
• Prompt-Architect hygiene: capture HARD/SOFT/INFERRED constraints (auth, rate limits, payload schema), ensure English-only output, and declare ceilings.
• Hook safety: validate signatures, enforce idempotency, set timeouts/backoff, and define circuit breakers; honor registry-only agents and latency budgets.
• Adversarial validation: fuzz payloads, test retries, replay, and timeout scenarios; capture evidence.
• MCP tagging: store hook playbooks under WHO=hooks-automation-{session} and WHY=skill-execution.

Execution Playbook

1. Intent & constraints: define hook purpose, endpoints, SLAs, and compliance; confirm inferred requirements.
2. Design: map triggers to actions, schema validation, auth methods, and telemetry.
3. Safety nets: configure retries/backoff, idempotency keys, and failure isolation with rollback.
4. Validation loop: fuzz and negative tests, replay/timeout drills, and timing checks; log evidence.
5. Rollout: stage deployment, monitor early signals, and prepare rollback.
6. Delivery: summarize design, evidence, risks, and confidence ceiling.

Output Format

• Hook intent, triggers, endpoints, and schemas.
• Auth/idempotency controls, retries/backoff, and rollback plan.
• Validation evidence (fuzz, replay, timeout) and risk register.
• Confidence: X.XX (ceiling: TYPE Y.YY) - rationale.

Validation Checklist

• Structure-first assets present or ticketed; examples/tests reflect hook cases.
• Auth, retries, and rollback validated; registry and hooks within latency budgets.
• Adversarial/COV runs logged with MCP tags; confidence ceiling declared; English-only output.

Completion Definition

Hook automation is complete when validations pass, monitoring is in place, risks are owned, and evidence is persisted with MCP tags.

Confidence: 0.70 (ceiling: inference 0.70) - Hook automation doc rebuilt with skill-forge scaffolding and prompt-architect confidence and constraint discipline.