Skip to content

ilandahan/AID / skills/reflection

reflection

Verified

Automatic quality check on significant outputs via isolated sub-agent. Spawns reflection-agent for objective evaluation without conversation context bias.

criteria/phase-1-prd.yaml

View on GitHub
# Phase 1: PRD - Reflection Criteria
# Product Requirements Document phase

phase: 1
name: "PRD"
description: "Requirements, user stories, acceptance criteria, scope definition"

criteria:
  - id: why_alignment
    name: "WHY Alignment"
    weight: 3
    checks:
      - "Does every requirement trace back to user/business need?"
      - "Is the WHY explicitly stated for each feature?"
      - "Are we solving the validated problem from Discovery?"
    scoring:
      excellent: "Clear WHY traceability for all requirements"
      acceptable: "Most requirements have clear WHY"
      poor: "Features without clear purpose or need"

  - id: phase_compliance
    name: "Phase Compliance"
    weight: 2
    checks:
      - "Defining WHAT, not HOW?"
      - "Not making technical/architecture decisions?"
      - "Not writing code or pseudocode?"
    scoring:
      excellent: "Pure requirements, no implementation details"
      acceptable: "Mostly requirements with minor tech hints"
      poor: "Contains architecture decisions or code"

  - id: correctness
    name: "Requirements Quality"
    weight: 3
    checks:
      - "User stories follow who/what/why format?"
      - "Acceptance criteria are testable and measurable?"
      - "Requirements are unambiguous?"
      - "No conflicting requirements?"
    scoring:
      excellent: "Clear, testable, unambiguous requirements"
      acceptable: "Good requirements with minor clarity issues"
      poor: "Vague, untestable, or conflicting requirements"

  - id: security
    name: "Security Requirements"
    weight: 2
    checks:
      - "Security requirements explicitly defined?"
      - "Data privacy considerations included?"
      - "Authentication/authorization needs specified?"
    scoring:
      excellent: "Comprehensive security requirements"
      acceptable: "Basic security requirements present"
      poor: "Security requirements missing or incomplete"

  - id: completeness
    name: "Scope Coverage"
    weight: 2
    checks:
      - "All user personas covered?"
      - "Happy path and error scenarios defined?"
      - "Edge cases identified?"
      - "Out of scope clearly stated?"
    scoring:
      excellent: "Complete coverage with clear boundaries"
      acceptable: "Main scenarios covered, some gaps"
      poor: "Major scenarios or personas missing"

threshold: 7
max_iterations: 3

notes:
  borderline: "Review acceptance criteria for testability"
  improvement_focus:
    - "Add explicit WHY to each requirement"
    - "Make acceptance criteria measurable"
    - "Clarify scope boundaries"
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall