Mechanical constraints for AI agents. Ship faster without the chaos.
Intense-Visions/harness-engineering
Command Claude Code
Pre-execution LLM-judgment skill: does a spec carry measurable, testable, complete acceptance criteria before work begins? Resolves the spec's acceptance section, critiques observability / testability / completeness (advisory), flags user-visible behaviors with no covering test (advisory), and emits a confidence-rated…
Intense-Visions/harness-engineering
Command Claude Code
Add a component to an existing harness project.
Intense-Visions/harness-engineering
Command Claude Code
Autonomous decide-stage orchestrator — sweep the backlog of pending architectural decisions, fan out worktree-isolated subagents that each run the real architecture-advisor pipeline to draft one ADR, independently verify every draft is a well-formed record, and hand the human one batch sign-off pass. Never…
Intense-Visions/harness-engineering
Command Claude Code
Apply codemods for safe DRIFT-T001/T002/T003 token-bypass findings; emit precise suggestions for DRIFT-T004 (deprecated tokens) and all DRIFT-P (primitive adoption). FIX half of the design-pipeline drift-remediation sub-project; pairs with detect-design-drift.
Intense-Visions/harness-engineering
Command Claude Code
LLM-judgment critique of API design quality — the ceiling counterpart to rule-based API checks (OpenAPI-format and webhook-format compliance). Asks whether resources model the domain rather than the implementation, whether resource naming and URL structure are predictable (path vs query param), whether HTTP methods…
Intense-Visions/harness-engineering
Command Claude Code
Interactive architecture advisor that surfaces trade-offs and helps humans choose.
Intense-Visions/harness-engineering
Command Claude Code
Rule-based brand-semantics audit. Detects token misuse (BRAND-T001 via $extensions.harness.brand.forbiddencontexts) and voice violations (BRAND-V001 via DESIGN.md voice.forbiddenphrases). 4th composed verifier in harness check-design. Triggers extraction of the formal verifier interface.
Intense-Visions/harness-engineering
Command Claude Code
Audit component definitions for missing required anatomy parts (slots, states, sizes) and detect missing-anatomy-component patterns (data without empty states, async without loading boundaries). First programmatic enforcer of component-anatomy rules.
Intense-Visions/harness-engineering
Command Claude Code
Mechanically audit a project's own harness setup against the seven STRENGTH failure patterns; reports per-pattern findings, a 0-100 strength score, and a tier label (solid/at-risk/theatre). Orchestrates harness check-harness-strength; never reimplements detection.
Intense-Visions/harness-engineering
Command Claude Code
Cross-dimensional codebase audit orchestrator — classify repo shape, fan out parallel read-only audit agents, dedup and severity-rank findings against existing issues, publish grouped tracking issues.
Intense-Visions/harness-engineering
Command Claude Code
Autonomous phase execution loop — chains planning, execution, verification, and review, pausing only at human decision points.
Intense-Visions/harness-engineering
Command Claude Code
Structured ideation and exploration with harness methodology.
Intense-Visions/harness-engineering
Command Claude Code
Proactive undiscovered-bug hunt across the standing codebase — rank the codebase into disjoint risk-ordered areas by composing the existing detection analyses, confirm the batch once, fan out worktree-isolated subagents that each run the real per-area hunt (review machinery, adversarial refutation, a tdd-authored…
Intense-Visions/harness-engineering
Command Claude Code
Install, calibrate, verify and diagnose a local Claude Code usage-pace HUD that cannot fail green.
Intense-Visions/harness-engineering
Command Claude Code
Monthly retrospective over skill-adoption telemetry — ranks most-invoked, failing, and abandoned-mid-workflow skills, flags stale ones, and reports catalog telemetry coverage.
Intense-Visions/harness-engineering
Command Claude Code
Autonomous CI/CD-remediation orchestrator — triage the red CI/CD-run and flaky-test backlog by cause, fan out worktree-isolated subagents that run the real deflake/heal pipeline, independently verify each fix by artifact and deterministic all-OS CI, and hand back a batch of remediation PRs for one bulk human review.…
Intense-Visions/harness-engineering
Command Claude Code
Detect and auto-fix dead code including dead exports, commented-out code, and orphaned dependencies.
Intense-Visions/harness-engineering
Command Claude Code
Autonomous entropy/hotspot remediation sweep — enumerate the entropy/hotspot backlog by composing the existing detection skills, rank the targets, confirm the batch once, fan out worktree-isolated subagents that each run the real per-target cleanup pipeline, independently verify each result by convergence artifact and…
Intense-Visions/harness-engineering
Command Claude Code
LLM-judgment critique of command-line ergonomics quality — the ceiling counterpart to mechanical CLI checks, and the one craft skill with no rule-based floor twin. Asks whether command and flag names are predictable and consistent, whether help text is task-oriented, whether errors are actionable, whether defaults are…
Intense-Visions/harness-engineering
Command Claude Code
LLM-judgment critique of code quality / readability (TS/JS source) — the ceiling counterpart to the rule-based code floor (entropy-cleaner for dead code / drift, enforce-architecture for boundaries + deps, complexity thresholds). Asks whether the code reveals intent and reads in the domain's language, whether the…
Intense-Visions/harness-engineering
Command Claude Code
Multi-phase code review pipeline with mechanical checks, graph-scoped context, and parallel review agents.
Intense-Visions/harness-engineering
Command Claude Code
Orchestrate dead code removal and architecture violation fixes with shared convergence loop.
Intense-Visions/harness-engineering
Command Claude Code
5-phase post-mortem capture. Writes a structured solution doc at docs/solutions/{track}/{category}/{slug}.md with frontmatter, overlap-detection, and per-category lock for concurrency safety.
Intense-Visions/harness-engineering
Command Claude Code
LLM-judgment critique of prose-in-code across six surfaces (error messages, log lines, CLI output, commit subjects, PR descriptions, code comments). Third craft-pipeline ceiling skill; primary domain is error messages (universally bad). Graceful degradation when git/gh prerequisites are absent.