hm-l2-spec-driven-authoring
Use when turning a PRD, specification, contract, or acceptance brief into falsifiable requirements, acceptance criteria, verification methods, or…
Skill by shynlee04
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/shynlee04-hm-l2-spec-driven-authoring
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents, testing, security
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Shell scripts
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/hm-l2-spec-driven-authoring
What's inside
- Skill name
hm-l2-spec-driven-authoring- Description
- Use when turning a PRD, specification, contract, or acceptance brief into falsifiable requirements, acceptance criteria, verification methods, or implementation-compliance checks. Trigger for spec-locking and requirement extraction. NOT for exploratory coding, generic planning, manual-only QA, or RED/GREEN/REFACTOR execution after requirements are already locked.
- License
- MIT
- Metadata
- role: domain-execution · layer: 2 · pattern: P2 · version: 1.2.0
- Instructions
- 220 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Agent Skills spec issues
- allowed-tools: must be a space-separated string
Files (15)
- SKILL.md
- evals.json · evals/
- .gitkeep · metrics/
- rich-eval-rubric.json · metrics/
- rich-gate-scorecard.md · metrics/
- acceptance-test-patterns.md · references/
- source-synthesis.md · references/
- spec-to-req-mapping.md · references/
- validate-rich-package.sh · scripts/
- validate-skill.sh · scripts/
- task_plan.md
- .gitkeep · templates/
- requirement-traceability-matrix.md · templates/
- .gitkeep · workflows/
- spec-lock-workflow.md · workflows/
SKILL.md (first lines)
# Spec-Driven Authoring
## Overview
Turn written intent into a falsifiable implementation contract. This skill owns spec intake, ambiguity rejection, requirement extraction, acceptance-test derivation, and handoff evidence. It does **not** execute TDD cycles; once requirements and acceptance tests are locked, hand off to `hm-test-driven-execution`.
This package synthesizes three inspected third-party patterns:
| Source pattern | Adopt / adapt decision | Local transformation |
|---|---|---|
| `addyosmani/agent-skills@spec-driven-development` | Adopt the spec-first gate sequence and assumption-surfacing discipline; adapt away from implementation ownership. | Use Source Audit → Ambiguity Gate → Requirement Table → Acceptance Matrix → Handoff Packet. |
| `proffesor-for-testing/agentic-qe@qe-requirements-validation` | Adopt SMART validation, traceability matrix, and BDD coverage framing; adapt away from AQE-specific runtime calls. | Use portable requirement-to-test matrices with project-local commands only. |
| `kw12121212/auto-spec-driven@spec-driven-sync-specs` | Adapt drift/mapping metadata; defer `.spec-driven/` layout assumptions. | Preserve optional implementation/test mapping metadata without requiring any planning framework. |
## Entry Gate
Proceed only when at least one source artifact exists: PRD, SPEC, ADR, user story, API contract, acceptance brief, regulatory requirement, or explicit user request to create falsifiable requirements.
Before authoring:
1. Identify the source artifact and intended reader.
2. Run `prompt-skim` when available to detect length, URLs, missing files, and candidate scope.
3. Run `prompt-analyze` when available to detect contradiction, vagueness, missing actors, missing success criteria, and hidden scope.
4. If either tool is unavailable, perform the same checks manually and state that the tool was unavailable.
5. Block exploratory implementation requests that do not contain a stable requirement source.
## Boundary Rules
| Nearby workflow | Boundary |
|---|---|
| `hm-test-driven-execution` | Consumes locked requirements and executes RED/GREEN/REFACTOR. This skill derives the requirements and expected tests only. |
| `hm-planning-persistence` | Persists multi-session plan/progress state in `.hivemind/state/planning/<session-id>/`. This skill may produce requirements tables for that state but does not own task scheduling. |
| Generic planning | Plans may sequence work. This skill locks what must be true and how to verify it. |
| Exploratory coding/prototyping | Not enough source truth. Return blocked or ask for a spec before deriving REQ-* items. |
## Spec-Lock Workflow
Use this default sequence. Do not skip gates because the source is short.
```text
Source Audit → Ambiguity Gate → Requirement Table → Acceptance Test Matrix → Handoff Packet
```
### 1. Source Audit
Create an audit note with:
- Source path or pasted source title.
- Date/time of review.
- Stakeholders or actors named by the source.
- Explicit MUST/SHOULD/MAY statements.
- Implicit requirements inferred from security, data integrity, accessibility, or operational correctness.
- Out-of-scope statements.
Record provenance in the format from `templates/requirement-traceability-matrix.md`. Every later REQ row must point back to a source quote, source section, or blocked missing source.
### 2. Ambiguity Gate
Reject or clarify any candidate requirement containing unbounded terms such as “fast”, “good”, “simple”, “secure”, “robust”, “intuitive”, “soon”, or “scalable” unless the source gives measurable thresholds.
If ambiguity remains after one rewrite attempt, return a blocked handoff instead of inventing precision.
Use SMART screening from the requirements-validation source pattern:
| Check | Pass condition | Block condition |
|---|---|---|
| Specific | Actor, action, object, and context are named. | Actor or object is implied but not stated. |
| Measurable | Observable output, state, event, metric, or error exists. | Outcome is subjective or unverifiable. |
| Achievable | Project constraints do not contradict it. | Source requires unsupported platform/tooling without adapter. |
| Relevant | It maps to the requested product/workflow goal. | It is implementation detail with no user/system outcome. |
| Testable | At least one verification method exists. | No automated, inspection, or manual verification can prove it. |
### 3. Requirement Extraction
For every requirement, write:
```markdown
### REQ-<domain>-<nn>: <short name>
**Source:** <quote or source section>
**Condition:** <exactly one falsifiable statement>
**Acceptance Criteria:**
- Given <state>, when <action>, then <observable result>.
**Verification Method:** <test, inspection, command, manual validation, or blocked reason>
**Integration Notes:** <agents/commands/tools/hooks/runtime-state implications>
**Status:** draft | locked | blocked
```
### 4. Acceptance-Test Derivation
For each locked requirement, derive:
- Positive case: expected success path.
- Negative case: validation, authorization, error, or invalid-input path when applicable.
- Boundary case: empty, minimum, maximum, missing dependency, timeout, or platform edge.
- Integration case: cross-component behavior when more than one runtime surface is involved.
Write the matrix before handoff:
```markdown
| REQ ID | Source quote/path | Positive | Negative | Boundary | Integration | Verification command/method | Coverage state |
|---|---|---|---|---|---|---|
```
`Coverage state` is `mapped`, `gap`, `blocked`, or `not-applicable`. A `gap` is acceptable only when it is visible in the handoff packet.
### 5. Handoff
When requirements are locked, provide a handoff packet for `hm-test-driven-execution`:
```yaml
requirements:
- id: REQ-...
condition: ...
acceptance_tests: [...]
red_phase_expectation: tests fail before implementation
verification_commands: [...]
blocked_items: [...]
traceability_matrix: path-or-inline-table
source_synthesis: references/source-synthesis.md decisions appliedFrom shynlee04/hivemind-plugin, MIT.
More from shynlee04/hivemind-plugin
- 0.578score
hm-l3-synthesis
shynlee04
Compress research findings into actionable artifacts with tiered reduction.
SkillCFreshness: C★ 51Claude Code +9 - 0.578score
hm-l2-coordinating-loop
shynlee04
Coordinate multi-agent dispatch with validation gates and handoff protocols.
SkillCFreshness: C★ 51Claude Code +9 - 0.578score
hm-l3-integration-contracts
shynlee04
Formal bidirectional contract authority between skills and agents.
SkillCFreshness: C★ 51Claude Code +9 - 0.578score
hf-l2-agent-composition
shynlee04
Teaches agents how to compose specialist agents from user intent using XML markup grammar, step protocols, and structured return formats.
SkillCFreshness: C★ 51Claude Code +9 - 0.578score
hf-l2-skill-synthesis
shynlee04
Use when synthesizing skills from GitHub repos, classifying skill patterns, building template libraries, generating eval frameworks, or creating eval-driven…
SkillCFreshness: C★ 51Claude Code +9 - 0.578score
hm-l3-detective
shynlee04
Investigate codebases with SCAN, READ, and DEEP reading modes.
SkillCFreshness: C★ 51Claude Code +9 - 0.574score
hf-l2-command-parser
shynlee04
This skill should be used only when parsing $ARGUMENT propositional commands from OpenCode command strings.
SkillCFreshness: C★ 51Claude Code +9 - 0.574score
hf-l2-skill-router
shynlee04
Maps meta-builder task domains to hf-* skill loading bundles with priority ordering.
SkillCFreshness: C★ 51Claude Code +9 - 0.574score
gate-l3-lifecycle-integration
shynlee04
Internal quality gate that evaluates whether Hivemind harness implementations correctly participate in the runtime lifecycle — covering 9-surface mutation…
SkillCFreshness: C★ 51Claude Code +9 - 0.574score
gate-l3-spec-compliance
shynlee04
Spec compliance gate performing bidirectional traceability, gap detection (4 types), EARS acceptance criteria validation, and anti-pattern scanning.
SkillCFreshness: C★ 51Claude Code +9 - 0.574score
hf-l2-delegation-gates
shynlee04
Enforce pre-delegation authorization gates before agent dispatch.
SkillCFreshness: C★ 51Claude Code +9 - 0.574score
hm-l2-requirements-analysis
shynlee04
Formal requirements diagnosis methodology for detecting gaps, contradictions, missing constraints, and unvalidated assumptions in existing requirements.
SkillCFreshness: C★ 51Claude Code +9
Signals
- Stars
- 51
- Forks
- 16
- Open issues
- 21
- Last commit (this folder)
- May 7, 2026 (5 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- Version
- 1.2.0
- License
- MIT
- Repository
- github.com/shynlee04/hivemind-plugin
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from shynlee04
- 0.578score
hm-l3-detective
shynlee04
Investigate codebases with SCAN, READ, and DEEP reading modes.
SkillCFreshness: C★ 51Claude Code +9 - 0.578score
hm-l3-integration-contracts
shynlee04
Formal bidirectional contract authority between skills and agents.
SkillCFreshness: C★ 51Claude Code +9 - 0.578score
hf-l2-skill-synthesis
shynlee04
Use when synthesizing skills from GitHub repos, classifying skill patterns, building template libraries, generating eval frameworks, or creating eval-driven…
SkillCFreshness: C★ 51Claude Code +9 - 0.578score
hm-l2-coordinating-loop
shynlee04
Coordinate multi-agent dispatch with validation gates and handoff protocols.
SkillCFreshness: C★ 51Claude Code +9 - 0.578score
hf-l2-agent-composition
shynlee04
Teaches agents how to compose specialist agents from user intent using XML markup grammar, step protocols, and structured return formats.
SkillCFreshness: C★ 51Claude Code +9 - 0.578score
hm-l3-synthesis
shynlee04
Compress research findings into actionable artifacts with tiered reduction.
SkillCFreshness: C★ 51Claude Code +9
Cite this page
hm-l2-spec-driven-authoring skill. Agent Mart. https://agentmart.dev/skills/shynlee04-hm-l2-spec-driven-authoring (last updated 2026-10-24)