aidlc-build
Validates that your code compiles, passes all test suites, and meets project quality standards.
Install
mkdir -p .claude/skills/aidlc-build && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/11787" && unzip -o skill.zip -d .claude/skills/aidlc-build && rm skill.zipInstalls to .claude/skills/aidlc-build
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Final integration build and test verification. Validates that implemented code compiles, passes all test suites, and meets quality gates before deployment.Key capabilities
- →Run full build processes
- →Execute complete test suites
- →Check code quality gates
- →Produce build artifacts
- →Detect build tooling from project configuration
- →Generate a build report
How it works
The skill detects build tooling and test frameworks from project configurations, then runs full builds and test suites. It checks quality gates like coverage and linting, and produces a build report.
Inputs & outputs
When to use aidlc-build
- →Run full build and integration test suites
- →Validate code against quality gates
- →Check test coverage thresholds
- →Generate a final build verification report
About this skill
Build Skill
Base:
shared/base.md(full on first load, §Summary on chain). Actions: load per-step fromactions/.
You verify that the implemented code is production-ready. Run builds, execute test suites, check quality gates, and produce build artifacts. You don't write new features — you validate what's been built.
When active:
- Follow ONLY the process below
- WAIT for user approval at each checkpoint
- Never narrate your internal process
- ALL output in the user's language (read manifest
languagefield) — no English narration
Activation
✅ aidlc-build active — {platform} detected.
Ready to verify build and run integration tests.
Quick Start
- Detect build tooling and test frameworks from project configuration
- Run full build → report results
- Run full test suite (unit, integration, E2E) → report results
- Check quality gates (coverage thresholds, lint, type-check, security scan)
- Produce build report → wait for approval → hand off to deploy
Reads: package.json / Makefile / build configs, tasks.md (for context), design/testing-strategy.md, blueprints Writes: build-report.md
Information Contract
Required Inputs
| Information | Description | Accepted Formats |
|---|---|---|
| Source code | Implemented code to verify | Filesystem access |
| Build configuration | Project build tooling (package.json, Makefile, Cargo.toml, etc.) | Config files in project root |
Optional Inputs
| Information | Description | Accepted Formats |
|---|---|---|
| Testing strategy | Expected coverage, test types, frameworks | Markdown (design/testing-strategy.md) |
| Quality thresholds | Coverage minimums, lint rules, security policies | Config files or steering |
| Design documents | Architecture context for integration verification | Markdown (design/*.md) |
Outputs
| Artifact | Default Path | Description |
|---|---|---|
| build-report.md | {WORKFLOW_DIR}/{feature}/build-report.md | Build results, test results, quality gate status |
Incremental Mode
- Scope: Run full project build (all units must integrate)
- Report at:
{WORKFLOW_DIR}/{feature}/build-report.md(project-wide, not per-unit)
Initialization
- Detect environment (per shared base)
- Resolve feature name (per shared base)
- Read manifest at
{WORKFLOW_DIR}/{feature}/aidlc-manifest.yaml - Detect build tooling:
- Scan for:
package.json,Makefile,Cargo.toml,pom.xml,build.gradle,pyproject.toml,Dockerfile,docker-compose.yml - Identify: build commands, test commands, lint commands, type-check commands
- Scan for:
- Read
design/testing-strategy.mdif exists (for coverage expectations) - Read blueprints Summary sections
Process
Execute actions sequentially. Load the action file when you reach that step — not before.
| Step | Action | Load |
|---|---|---|
| 1 | Detect and confirm build configuration | {SKILL_DIR}/actions/detect.md |
| 2 | Run build and tests | {SKILL_DIR}/actions/verify.md |
| 3 | Generate build report | {SKILL_DIR}/actions/report.md |
Skill Handoff
Next skill: aidlc-deploy (on user approval of build report).
Phase-Specific Rules
- This is a project-wide phase — always runs at the feature level, never per-unit.
- For incremental mode: all units must be implemented before this phase runs.
- Audit actions: build-detect, build-run, test-run, quality-check, build-approved.
Build Rules
- Run the full build exactly as CI would — no shortcuts
- All tests must pass (unit, integration, E2E if configured)
- Quality gates are advisory by default — report failures but let the user decide
- If build or tests fail, present diagnostics and offer to fix or skip
- Never modify source code in this phase unless explicitly asked to fix a failing test/build
Quality Gates (check if configured)
- Test coverage meets threshold (from testing-strategy.md or config)
- No lint errors (if linter configured)
- Type-check passes (if TypeScript, Flow, mypy, etc.)
- No critical security vulnerabilities (if scanner configured)
- Build produces expected artifacts (bundles, binaries, images)
Context Recovery
If context is lost mid-phase, follow aidlc/shared/base.md Context Recovery, then (the manifest is the state authority, not the report header):
- No
build-report.mdat{WORKFLOW_DIR}/{feature}/→ loadactions/detect.md(start from detection) - Report exists + manifest
artifacts.build.statusis"draft"or missing → loadactions/report.md, re-present for approval (step 3) - Manifest
artifacts.build.statusis"approved"/"approved-with-warnings"→ build is complete; hand off toaidlc-deploy
When not to use it
- →When writing new features
- →When the task is not about validating existing code
- →When the phase is not project-wide
Limitations
- →Quality gates are advisory by default
- →Does not modify source code unless explicitly asked to fix a failing test/build
- →Always runs at the feature level, never per-unit
How it compares
This skill automates the entire build and test verification process, unlike manual execution of individual build and test commands.
Compared to similar skills
aidlc-build side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| aidlc-build (this skill) | 0 | 1mo | No flags | Intermediate |
| agent-production-validator | 3 | 6mo | Review | Advanced |
| validate-delivery | 1 | 5mo | Review | Beginner |
| release-testing | 1 | 1mo | Review | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by aws-samples
View all by aws-samples →You might also like
agent-production-validator
ruvnet
Agent skill for production-validator - invoke with $agent-production-validator
validate-delivery
avifenesh
Use when validating task completion before shipping. Runs tests, build, and requirement checks. Returns pass/fail with fix instructions.
release-testing
mono
Run integration tests to verify SkiaSharp NuGet packages work correctly before publishing. Use when user asks to: - Test/verify packages before release - Run integration tests - Test on specific device (iPad, iPhone, Android emulator, Mac, Windows) - Verify SkiaSharp rendering works - Check if packages are ready for publishing - Run smoke/console/blazor/maui tests - Continue with release - Test version X Triggers: "test the release", "verify packages", "run tests on iPad", "check ios tests", "test mac catalyst", "run android tests", "continue", "test 3.119.2-preview.2".
documenso-ci-integration
jeremylongshore
Configure CI/CD pipelines for Documenso integrations. Use when setting up automated testing, deployment pipelines, or continuous integration for Documenso projects. Trigger with phrases like "documenso CI", "documenso GitHub Actions", "documenso pipeline", "documenso automated testing".
smoke-check
hoatv2211
Run core path smoke validation before QA handoff or merge.
release-bump
himatts
Prepare versioned release updates for Lime Pipeline. Use when user-visible behavior changes require bumping `bl_info["version"]`, updating `CHANGELOG.md`, and producing release-ready QA notes.