Skill Claude CodeCodex
Scaffold a Spring Data JPA @Entity correct-by-default so the guardrails never fire: @Version optimistic locking, business-key equals/hashCode (never the generated @Id), JPA auditing fields, optional soft-delete, plus a matching repository — all matching this project's conventions. Trigger phrases: "create entity"…
Skill Claude CodeCodex
Scaffold JWT bearer authentication for a stateless JSON API. Recommends Spring Security's resource server (framework verifies signature + exp); also ships a custom OncePerRequestFilter using jjwt for teams that need it — it verifies the signature, rejects alg=none, checks exp/nbf, and returns 401 JSON on failure.…
Skill Claude CodeCodex
Scaffold a global REST error contract — a @RestControllerAdvice that maps every exception to an RFC 7807 ProblemDetail (Spring 6): validation errors → 400 with field details, a domain NotFoundException → 404, and a fallback 500 that never leaks stack traces. This is the fix the entity-in-controller / error-contract…
Skill Claude CodeCodex
Scaffold a Spring Security 6 SecurityFilterChain @Bean baseline — deny-by-default authorizeHttpRequests, a BCryptPasswordEncoder, explicit CORS, and CSRF guidance (kept for cookie/session auth, disabled only for stateless token APIs). Lambda DSL, Jakarta, no deprecated .and() chaining. Trigger phrases: "set up spring…
Skill Claude CodeCodex
Wire the Spring Boot guardrails into this project — install the jpa-guardrail PostToolUse hook + ruleset, register it in settings.json, copy the springboot-reviewer agent, and inject the Spring Boot rule block into CLAUDE.md. Short interview fills project slots (base entity, migration tool, strict mode). Trigger…
Skill Claude CodeCodex
Scaffold a Spring Web REST endpoint correct-by-default — controller + request/ response DTOs + service method + mapper. Never exposes a JPA entity (satisfies the entity-in-controller guardrail). Request bodies are @Valid DTOs. Trigger phrases: "add endpoint", "scaffold rest controller", "new REST endpoint", "create…
Skill Claude CodeCodex
Fine-grained Spring Boot 3.x + JPA conventions and the rationale behind the toolkit's guardrail rules. Split by domain — load the reference for what you're touching. Use when working on persistence, controllers, transactions, or tests and you want the why, not just the what. Trigger phrases: "spring boot conventions"…
Skill Claude CodeCodex
Scaffold focused Spring slice tests — @DataJpaTest for the persistence layer (rolls back per test) and @WebMvcTest for the web layer (MockMvc + mocked services). Pick the narrowest slice that proves the behavior; reach for full @SpringBootTest only for cross-layer flows. Trigger phrases: "datajpatest", "webmvctest"…
Plugin Claude Code
Self-improving Compose Multiplatform UI harness for Claude Code. A render -> grade -> ratchet loop: inspection tests render every screen across a stress matrix, a deterministic Tier-1 asserter plus a vision evaluator grade them in a separate lane, and recurring findings ratchet into permanent gates via a compounding…
Agent
Grades rendered mobile UI screenshots against the mobile-design rubric and returns a pass/fail verdict with element-level fixes. Dispatch it AFTER an inspection harness has rendered a screen's PNGs (e.g. SongsScreenInspection → build/outputs/roborazzi/inspect.png), especially after any @Composable edit, to close the…
Command
Distill the feedback ledger into promotion proposals and — on human approval — harden recurring findings into permanent gates, then delete the now-redundant soft guidance. The self-improvement ratchet's act-half.
Command
Iterate on a Compose screen until it passes the structural gate + visual evaluator — renders the stress matrix, grades it in a separate lane, logs findings to the ledger. Caps at 4 iterations then escalates.
Command
Render the harness's self-improvement metrics from the ledger — iterations-to-pass, first-pass yield, and recurrence-after-deposit — to show whether the harness is actually compounding.
Skill Claude CodeCodex
Use when adding or extending automated module-boundary / architecture enforcement in a Kotlin or Kotlin Multiplatform Gradle project — rules like "core must not depend on feature", "api must not leak impl", layer/package dependency directions that should fail the build the moment a bad import is written. Covers tool…
Skill Claude CodeCodex
Evaluation criteria for grading rendered mobile UI screens. Use when reviewing, critiquing, or self-checking a mobile UI screenshot — especially after any UI edit in an agentic loop — to decide pass/fail and produce concrete fixes. Targets Compose Multiplatform (Android + iOS), Material 3. The mobile counterpart to…
Plugin Claude Code
Sets up a standardized Claude Code harness for your project: CLAUDE.md, permissions, hooks, MCP servers, agents, SSOT docs, and observability logging. One init interview, consistent harness across every team member.
Agent
Analyze harness usage patterns and suggest improvements based on actual tool call logs. Reads .claude/logs/ and surfaces actionable hook and gate suggestions. Trigger phrases: "optimize harness", "harness suggestions", "improve my setup", "what hooks should I add", "analyze tool usage", "what should I change in my…
Command
Create a new Architecture Decision Record in docs/adr/. Quick operation. Trigger phrases: "create ADR", "document decision", "record architecture choice", "add ADR", "document this decision".
Command
Create a new agent in .claude/agents/. Quick operation, no full interview needed. Trigger phrases: "add agent", "create agent", "add [name] agent".
Command
Add a new hook to .claude/settings.json. Quick operation, no full interview needed. Trigger phrases: "add hook", "auto-format on save", "run tests after edit", "lint on write", "run [command] after [tool]".
Command
Add a new MCP server to .mcp.json. Quick operation, no full interview needed. Trigger phrases: "add MCP", "connect to [service]", "configure [service] MCP".
Skill Claude CodeCodex
Set up the Claude Code harness for this project. Configures CLAUDE.md, permissions, hooks, MCP servers, agents, and SSOT docs based on your stack and chosen tier. Run on first setup or when CLAUDE.md is missing or has [PLACEHOLDER] markers. Trigger phrases: "set up harness", "configure claude", "onboard project"…
Skill Claude CodeCodex
Start a new project from scratch. 3-phase interview: project goals → tech stack → architecture → conventions. Scaffolds using official tools then configures the Claude Code harness. Run on an empty directory. Trigger phrases: "start a new project", "create project from scratch", "greenfield setup", "I have an empty…
Skill Claude CodeCodex
Analyze Claude Code tool call logs and suggest harness improvements based on actual usage patterns. Requires observability enabled + 7 days of data. Trigger phrases: "optimize harness", "harness suggestions", "improve my setup", "analyze tool usage", "what hooks should I add".