KienVS
Community@juartaurus98 · Vietnam
404 ~~ Not Found
Agent Skills by KienVS
Showing 15 vetted skills indexed across 1 GitHub repositories.
sdlc-init
Bootstraps the SDLC framework directory structure and project context files in a repository.
sdlc-trace
Audits traceability across specifications, code, and tests by recomputing scenario hashes and rebuilding the trace index.
sdlc-code
Implements use cases from technical designs following layered architecture and trace tagging rules.
sdlc-prd
Generates product requirements documents with numbered use cases and measurable success metrics from feature intake documents.
sdlc-kernel
Defines the shared protocol, phase graph, and conventions for the sdlc-* SDLC framework.
sdlc-srs
Generates software requirements specifications with executable Gherkin acceptance criteria from PRD documents.
sdlc-techdoc
Generates technical design documents with component maps, interface specs, test strategies, and ADRs from SRS inputs.
sdlc-testcase
Generates human-readable test case catalogs from SRS scenarios and tech design test strategies.
sdlc-unittest
Implements unit and mocked-integration tests from a test case catalog with traceability tags and coverage gates.
sdlc-eval
Runs golden-set evaluations for LLM prompt versions and compares quality metrics against baselines.
sdlc-review
Reviews code changes against specifications and project standards, then writes findings to state.
sdlc-loop
Routes the SDLC development loop by reporting feature status and the next command.
sdlc-check
Validates whether SDLC phase input documents are ready to start the next phase.
sdlc-feature
Converts a one-line feature request into a structured intake document with a FEAT-ID and state machine entry.
sdlc-testrun
Implements and executes acceptance and live-integration tests, then triages failures into routed defect categories.
Frequently Asked Questions About KienVS
FAQPage SchemaWhat tasks can I perform using KienVS's sdlc-* skills?▼
You can run the full delivery chain: capture feature requests (/feature), write PRDs (/prd) and SRS documents with Gherkin criteria (/srs), produce technical designs with ADRs (/techdoc), generate code (/gen-code), derive test catalogs (/gen-testcase), implement unit tests (/unittest), run acceptance tests (/run-test), review diffs (/review), and audit traceability (/trace).
Who are these sdlc-* skills designed for?▼
Software engineers, tech leads, and QA practitioners who want a gated, specification-first development chain. Each phase is invoked by an explicit slash command, and /check validates handoff readiness by naming missing items by file, section, and responsible role.
How do I install and start using the SDLC framework?▼
Run /init-project in your repository. The sdlc-init skill detects the stack, fills .agent/project-context.yaml, creates docs/ and .agent/state/ trees, writes the managed project-facts block into CLAUDE.md, and seeds domain knowledge files. It is idempotent and supports --upgrade for re-runs.
How does the framework handle AI-backed features?▼
The /feature intake classifies whether a feature is AI-backed. For those, /prd states the quality bar in user terms, /techdoc writes prompt and eval-set specs, /gen-testcase authors evaluation cases, and /eval runs a golden-set harness recording quality metrics and cost baselines to detect regression across prompt versions.
How do I track progress and diagnose problems mid-chain?▼
Run /sdlc to see where every feature stands, the single next command, and current blockers. Use --blast to preview regeneration scope of a spec change, --doctor to self-check the framework, and /fix-bug as a mid-chain entry point for defect routing.