Agent
Domain specialist for the Trabuco security audit — Data + Events. Loads checklist-data-events.md, walks the project source tree, and returns structured findings (one record per check that fired) to the orchestrator. Read-only; never modifies project files. Invoked by trabuco-security-audit-orchestrator via the Task…
Agent
Top-level orchestrator for the Trabuco security audit. Loads the canonical checklist from plugin/skills/audit/ (or the per-project copy in .ai/security-audit/ when running inside a generated project), dispatches five domain specialists in parallel via the Task tool (auth, ai-surface, aiagent-java, data-events…
Agent
Domain specialist for the Trabuco security audit — Web + Infra. Loads checklist-web-infra.md, walks the project source tree, and returns structured findings (one record per check that fired) to the orchestrator. Read-only; never modifies project files. Invoked by trabuco-security-audit-orchestrator via the Task tool…
Hook
Runs when a session starts, executing session-start.sh via bash. From arianlopezc/Trabuco.
Hook
Runs after a tool call finishes for mcp__trabuco__init_project and mcp__trabuco__generate_workspace tool calls, executing post-init-project.sh and post-generate-workspace.sh via bash (2 commands). From arianlopezc/Trabuco.
Skill Claude CodeCodex
Add a Trabuco module to an existing project. Detects current project state, recommends compatible modules, and wires them in. Use when the user has an existing Trabuco project and wants to extend it (add AIAgent, add EventConsumer, add NoSQL, etc.).
Skill Claude CodeCodex
Run the Trabuco security audit against the current Trabuco-generated project. Dispatches a multi-domain check across auth, AI surface, AIAgent runtime, data persistence + messaging, and web/infra layers using the canonical checklist bundled with this skill. Produces a severity-sorted findings report and an explicit…
Skill Claude CodeCodex
Decompose a multi-service system into independent Trabuco services. Use when the user is describing a system of multiple cooperating services (marketplace, platform, microservices architecture) rather than a single service.
Skill Claude CodeCodex
Validate a Trabuco project's structural health. Runs the Trabuco doctor + interprets findings. Use when a project isn't building correctly, after a manual edit, or before committing significant changes.
Skill Claude CodeCodex
Propose what to add next to an existing Trabuco project based on its current state and what the user wants to achieve. Different from /trabuco:add-module — that adds a specific module; this one ADVISES on what to add. Use when the user has a project and asks "what's next?" or "how would you extend this?".
Skill Claude CodeCodex
Migrate an existing Java repository in place into a Trabuco-shaped multi-module project. Drives the 14-phase orchestrated flow with specialized subagents, dependency-aware phasing (legacy CI keeps working at every phase boundary), per-phase approval gates, and atomic rollback via git tags. Use when the user has an…
Skill Claude CodeCodex
Start a new Java/Spring Boot project from natural-language requirements. Recommends modules, confirms with the user, then generates via the Trabuco CLI. Use when the user wants to create a new service and describes what it should do.
Skill Claude CodeCodex
Produce a stratified implementation plan for a multi-module change in a Trabuco-shaped Java/Spring Boot project. Each plan stage is one module in dependency order (Foundation → Contracts → Persistence → BusinessLogic → Edge), with per-stage validate command and depends-on rationale. Use when the user wants an explicit…
Skill Claude CodeCodex
Pure architecture advice — no generation. Given requirements (greenfield or brownfield), recommends module selections, patterns, and design tradeoffs grounded in Trabuco's catalog. Use when the user wants to THINK before committing, evaluate options, or understand what Trabuco would do for their problem.
Skill Claude CodeCodex
Bring an existing Trabuco project's AI tooling up to date with the installed CLI. Detects missing skills, subagents, prompts, hooks, and review scaffolding, then creates them additively. Never modifies existing files.