sdd-explore
Explore SDD ideas before committing to a change. Trigger: orchestrator launches exploration or requirement clarification.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-10
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- sdd-explore
- triggers
- SDD explore, explore ideas, investigation, discovery, SDD exploration
- changelog
- docs/ciclos/cycle28-20260815.md
- description
- Explore SDD ideas before committing to a change. Trigger: orchestrator launches exploration or requirement clarification.
- token_budget
- 2440
- delegate_only
- true
03 SKILL.md 正文
Executor Override
If you ARE the sdd-explore sub-agent, gate does not apply. Execute directly.
Language
Artifacts default to English. Spanish: neutral/professional unless regional variant requested.
Purpose
Explore codebase for a topic/feature, compare approaches, return structured analysis. Research only by default; create exploration.md only for named changes.
Input
- Topic or feature to explore
- Artifact store mode (
engram | openspec | hybrid | none)
Workflow
- Load Skills — §A from
sdd-phase-common.md - Understand Request — Parse: feature? bug fix? refactor? Domain?
- Investigate — Read entry points, search related code, check tests, identify patterns, map dependencies.
- Analyze Options — Compare approaches (table: Approach | Pros | Cons | Complexity: Low/Med/High).
- Persist Artifact — Mandatory for named changes. §C from
sdd-phase-common.md: artifactexplore, topic_keysdd/{change}/explore, typearchitecture. - Return Analysis (write to
exploration.mdif saving): Current State, Affected Areas, Approaches (table), Recommendation, Risks, Ready for Proposal.
Rules
- Only file you MAY create:
exploration.mdin change folder - DO NOT modify existing code
- ALWAYS read real code, never guess
- Keep CONCISE — summary, not novel
- State if info insufficient or request too vague
- Return envelope per §D from
sdd-phase-common.md
docs/skills/sdd-explore/reference.md
Anti-Rationalization
| Rationalization | Red Flag | Verification |
|---|---|---|
| "Skill without verification" | Doing work without checking output format | Output matches skill ## Output contract + file:line citaton |
| "Save time skipping this skill" | Using skill directly without resolving deps | skill-graph resolution + cross-ref check |
| "Output is self-evident" | No file:line or confidence marker | Cite file:line or flag confidence: unvalidated |
Red Flags
- Doing work without checking output format → STOP, re-read skill
- Second occurrence of same rationalization → force RED zone
Verification
- Output matches skill ## Output contract + file:line citaton
- cross-ref-check.ps1 → SKILL.md OK
Refs
Cross-Refs: sdd | sdd-propose
04 同仓库更多 Skills
web-quality-audit
Comprehensive web audit: performance, a11y, SEO, responsive, animation, design tokens.
ui-engine
UI system — Grid/Flexbox/@layer/:has(), container queries, compositor-only animation, OKLCH tokens, component patterns
baseline-ui
Anti-slop UI — layout, typography, responsive, animation, tokens. Use for cleanup or polish.
skill-creator
Create new AI agent skills following Agent Skills spec -- auto-bootstrap unknown tasks, define structure, register in AGENTS.md
senior-engineer
Apply senior engineering competencies — system design, trade-off analysis, delegation, mentoring, and tech debt strategy
prompt-engineering
Design effective prompts using SPEARS framework — scope, principles, examples, assertions, refinements, with ReAct and multi-agent patterns