| name | spec-driven-brief-planner |
| description | Use this skill when the Spec-Driven agent selects the brief lane for a bounded change that still requires a written plan and human approval before implementation. Default path: investigate targeted repository evidence, write and validate brief.md, stop for one approval, then implement the complete approved plan and run one compact review. |
Spec-Driven Brief Planner
Use one compact artifact and one approval for changes that do not require the full requirements, design, and tasks lifecycle.
Contract
- Write
.specs/changes/<slug>/brief.md before implementation.
- Do not write implementation code until the user explicitly approves the brief.
- Do not create
requirements.md, design.md, or tasks.md in the brief lane.
- If full-lane risk appears, stop and escalate before continuing.
Full-Lane Escalation
Escalate when any condition is true:
- security, authentication, authorization, privacy, secrets, or compliance changes
- persisted data, schema, migration, or persisted-format changes
- public API, external contract, wire-format, or backward-compatibility changes
- architecture changes across packages, services, or runtime boundaries
- new production dependency, deployment, rollback, or infrastructure work
- critical concurrency, availability, or performance behavior
- more than 5 requirements or more than 8 implementation tasks
Planning Workflow
- Propose a short URL-friendly slug.
- Read only repository guidance relevant to this change.
- Inspect targeted existing code before naming files or verification commands.
- Work inline by default. Use at most one explore sub-agent only when the investigation would otherwise require more than three independent searches. Do not repeat delegated searches.
- Write
.specs/changes/<slug>/brief.md using the exact format below.
- Run
sds validate structure <slug> and sds validate brief .specs/changes/<slug>/brief.md.
- Fix validation failures and re-run both commands.
- Stop and ask:
Verdict: PASS. Approve brief <slug> for implementation?
Brief Format
# Change Brief
**Lane:** brief
## Intent
<problem, expected outcome, and scope boundary>
## Non-Goals
- <explicitly excluded behavior>
## Acceptance Criteria
- REQ-1: <observable outcome>
- REQ-2: <observable outcome>
## Repository Evidence
- Existing: `path/to/file.ts` - <evidence and role in the change>
- New: `path/to/test.ts` - <planned role>
## Implementation Plan
- [ ] 1. <concrete action>
- Change: <exact scoped change>
- Files: `path/to/file.ts`
- Verify: `focused command or observable check`
- _Implements: REQ-1_
## Risks
-
## Risks is optional. Every other section is required and must stay in the shown order.
Brief Rules
- Use 1-5 sequential
REQ-N acceptance criteria.
- Use 1-8 sequential checkbox tasks.
- Keep acceptance criteria observable and implementation-independent.
- Give every task
Change, Files, Verify, and _Implements fields.
- Cover every requirement with at least one task.
- Use exact existing or proposed paths; do not leave placeholders.
- Do not use
DES-* IDs in the brief lane.
Approved Implementation
After explicit approval:
- Re-read the approved brief and changed working-tree state.
- Implement all pending tasks in order without asking between tasks.
- After a task's focused verification passes, mark it
[x] immediately.
- Do not re-read a file already read in the current session unless it was modified.
- Run every item under
Final Verification.
- Stop only for failed verification that cannot be fixed locally, material ambiguity, user conflict, or full-lane escalation.
Compact Review
Run exactly one independent compact review after implementation.
If a Task tool is available, launch one general sub-agent with:
Task: Brief Review for <slug>
Read:
- .specs/changes/<slug>/brief.md
- changed implementation files
Evidence:
- verification commands already run and their outcomes
Check:
1. Every REQ-N is implemented and verified.
2. The diff stays within Intent and Non-Goals.
3. File deviations are justified.
4. Tests exercise real behavior and are not skipped, hardcoded, or mocked past the behavior under test.
5. Reuse existing verification evidence; rerun only if it is absent or stale.
Return:
- Verdict: APPROVE | APPROVE WITH NOTES | REQUEST CHANGES
- Blocking findings: [list]
- Non-blocking findings: [list]
If no Task tool exists, perform the same review inline. Fix blocking findings and re-review once. Do not raise the acceptance bar or introduce unrelated findings in the second pass.
Quality Bar
Before requesting approval, verify: