Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Specification: [Title] — measured AI agent skill | DecimalAI

a5c-ai/spec-driven-development

SkillScore20Maintained

Specification creation and management for the Pilot Shell methodology. Covers semantic search, clarifying questions, structured spec generation, and iterative refinement.

by a5c-ai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Coding pack

decimalai skills pull a5c-ai-spec-driven-development

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/a5c-ai-spec-driven-development/SKILL.md

Raw file: /s/a5c-ai-spec-driven-development/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-spec-driven-development.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

25 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 9d ago

—-3% tokens0% turns20% of whole cases pass with skill (5/25)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-05	✗→✓	▲ Improved	28%	0%
case-08	✗→✓	▲ Improved	32%	0%
case-11	✗→✓	▲ Improved	-60%	0%
case-20	✗→✓	▲ Improved	86%	0%
case-22	✗→✓	▲ Improved	-41%	0%

Showing 5 of 25 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/31/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

a5c-ai / babysitter

library/methodologies/pilot-shell/skills/spec-driven-development/SKILL.md

View on GitHub ↗

License MITImported 7/31/2026Version v1

Skill contents

SKILL.md

Search for files and code related to the task description
Identify existing patterns that the spec should follow
Map the impact area of proposed changes
Generate SEARCH-CONTEXT.md with findings

2. Clarifying Question Resolution

Identify ambiguities in the task description
Generate targeted clarifying questions
Resolve assumptions with explicit choices
Document decisions for traceability

3. Spec Generation

Structure specs with: title, goals, tasks, acceptance criteria
Decompose into atomic, testable tasks
Define dependency graphs between tasks
Include rollback plans and risk assessments
Generate SPEC.md document

4. Iterative Refinement

Accept plan-reviewer feedback
Apply revision requests by severity
Refine task decomposition
Update acceptance criteria

Spec Structure

markdown
# Specification: [Title]

## Goals
- [ ] Goal 1 with measurable outcome

## Tasks
### Task 1: [Description]
- **Acceptance Criteria**: ...
- **Test Strategy**: RED->GREEN->REFACTOR
- **Complexity**: low/medium/high
- **Dependencies**: [task-ids]

## Assumptions
- Assumption 1 (validated: yes/no)

## Risks
- Risk 1: Mitigation strategy

Want to see if a5c-ai/spec-driven-development would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.