Run end-to-end quality gates and verify spec compliance from test receipts.
npx skills add https://github.com/Sungmin-Cho/claude-deep-work --skill deep-test
Please help me install this Agent Skill. Skill: deep-test Source: https://github.com/Sungmin-Cho/claude-deep-work/tree/main/skills/deep-test Command: npx skills add https://github.com/Sungmin-Cho/claude-deep-work --skill deep-test
Deep-test enforces rigorous, evidence-driven verification of implemented slices by running end-to-end test and quality gates, detecting design drift, and ensuring receipts demonstrate spec compliance so that delivered changes meet project expectations without premature completion claims.
Run the deep-test phase for the active session to execute all quality and verification gates, generate test-results.md and reports, and automatically retry failing slices within configured retry limits.
High-intent search queries and answers about installing and using this skill.
Comprehensive test verification runs end-to-end quality gates, validates receipt completeness, and detects design drift against baseline diffs to confirm implemented code slices meet spec compliance before project completion.
Drift detection determines a baseline commit, computes git diffs, and surfaces implemented versus not-implemented items to identify design drift and ensure delivered changes match original project specifications.
Automating retry orchestration for failing TDD cycles increments retry counters, moves session state back to implement for targeted fixes, and re-runs full verification up to configurable limits until all quality gates pass.
Yes, automated quality gates auto-detect existing project test commands and run sequential verification steps including cross-slice review, sensor clean, coverage, and mutation testing where available to produce structured reports and artifacts.
Yes, cross-slice reviews and quality gates require projects using the deep-work Brainsearch to Test workflow with per-slice TDD, access to session state and receipts, git history for baseline diffs, and evaluator models for automated analysis.
Test verification can fail despite passing individual test cases because the pipeline checks receipt completeness, validates TDD evidence per slice, runs cross-slice reviews, and detects design drift that individual tests may not surface.
TECHKNOWMAD-LABSAutomate a 7-phase software development lifecycle from brainstorming to merge.
cristoslcRun two-phase integration and smoke tests via swain-test.sh outputs.
Sungmin-ChoOrchestrate multi-phase software development workflows with TDD enforcement and git worktree support.
JaiminVadadoriyaAnalyze test coverage gaps in JavaScript, TypeScript, and Python applications.
