Claude Skills by GobbyAI

AST-aware code search, symbol navigation, and dependency graph analysis via the gcode CLI. Use this skill when exploring codebases, finding functions or classes, understanding call graphs, checking blast radius before changes, or navigating large projects. Triggers on: code search, find function, who calls, what imports, blast radius, symbol lookup, file outline, project structure, search codebase, find definition, impact analysis, dependency graph.

Internal methodology for the architecture discovery stage. Produces a concise Architecture Brief for product and planning handoff.

Anti-slop output mode. Kills filler, summary-stamps, negation framing, follow-up menus, and plain-language restatement. Three compression levels: lite, full, ultra.

Use when coordinating a full gobby build run for an epic or task, especially when the user assigns the current session as coordinator, asks for a coordination epic, wants build agents/worktrees monitored, or wants gobby build bugs fixed so future runs work unattended.

Use when user asks to 'build rule', 'create rule', 'author rule', 'write rule', 'add enforcement'. Interactive guide for authoring Gobby rule YAML definitions.

Interactive /gobby build wizard. Starts lifecycle automation from a plan file or task ref, or delegates raw ideas to /gobby plan.

Instructions for rendering rich A2UI canvas elements to the chat interface.

Instructions for using gcode CLI for code search and retrieval. Loaded on demand when project has a code index.

How to run code reviews and adversarial reviews. Use when asked to review code, run adversarial review, or when the code review gate blocks.

Use when processing CodeRabbit review comments, CodeRabbit CLI reports, or `$gobby coderabbit [findings]` requests.

Instructions for using context7 MCP to look up library documentation before writing code. Loaded on demand on first code file write/edit.

Interactive /gobby dev launcher. Dispatches the shared developer agent for a task.

Developer-agent discipline for test judgment, validation evidence, and TDD-required task escalation.

Select the best bundled developer agent for automated expansion leaves and document fallback decisions.

Review an implemented epic against its approved plan, aggregate diff, and child-task validation evidence before final PR or merge handoff.

Internal methodology for the ideation discovery stage. Produces a concise Discovery Brief for downstream research, architecture, PRD, and planning.

Create distinctive, production-grade frontend interfaces with high design quality. Generates creative, polished code that avoids generic AI aesthetics. Use when the user asks to build web components, pages, artifacts, posters, or applications, or when any design skill requires project context. Call with 'craft' for shape-then-build, 'teach' for design context setup, or 'extract' to pull reusable components and tokens into the design system.

How to discover and load skills. Covers local skill search, hub search, and when to proactively look for skills.

Methodology for PR-aware delivery — probe branch protection, open/gate PRs, land via GitHub when required, or run the direct multi-worktree merge campaign.

REQUIRED for all image generation requests. Generate and edit images using Nano Banana (Gemini CLI). Handles blog featured images, YouTube thumbnails, icons, diagrams, patterns, illustrations, photos, visual assets, graphics, artwork, pictures. Also routes specialty creative recipes (anime-to-life, character-reference-sheet, figure-to-life, imax-portrait, j-cover, j-idol, j-poses, photo-restoration, real-mecha). Use this skill whenever the user asks to create, generate, make, draw, design, or...

Switch the current session to a persona-capable agent definition without spawning a child agent or changing provider/model/isolation. Use with `/gobby persona <name>`.

Methodology for drafting a gobby plan document — phases, task format, TDD compatibility, categories, hierarchy, and dependency notation. Use when drafting or revising a plan artifact.

Review a gobby plan document for missing requirements, bad sequencing, unhandled edge cases, weak testability, and traceability gaps. Use when asked to review or critique a plan.

Internal methodology for the PRD discovery stage. Produces a concise Product Reference Document for implementation planning.

Enforces default Python coding standards for agents writing or refactoring Python: typing, error handling, testing, async, and boundary validation. Use before editing Python unless the repo provides stricter local rules.

Interactive /gobby qa launcher. Dispatches the shared qa-reviewer agent for a task.

Internal methodology for the research discovery stage. Produces concise evidence and source-backed context for architecture and PRD stages.

Interactive /gobby review launcher. Dispatches holistic-reviewer for an epic.

Enforces default Rust coding standards for agents writing or refactoring Rust: ownership, error handling, types, testing, async, and API design. Use before editing Rust unless the repo provides stricter local rules.

Commit message format and release PR process. Use when ready to commit or push a release; use task-transitions for task close/review gates.

How to create tasks effectively via gobby-tasks MCP. Covers task types, priorities, when to use each, and writing effective validation criteria.

Pre-transition checklist for closing or reviewing tasks. Covers fresh verification evidence, commits, and memory gates.

How to author, audit, and refresh Gobby technical documentation. Use when writing or rewriting a guide, validating an existing doc against the codebase, generating Mermaid diagrams, or explaining a technical concept to docs readers.

Internal methodology for architect-owned test architecture. Produces structured testing strategy prose inside the architecture discovery marker.

Required TDD loop for tasks that need red, green, refactor, final-green, and test-quality evidence.

Internal methodology for side-effect-free GitHub issue triage verdicts.

Use when creating, editing, or validating Gobby skills, especially when a skill must change agent behavior or become discoverable through gobby-skills.

Shorthand alias for /gobby. Usage: /g <skill> [args] or /g mcp <server> [args]

Unified router for Gobby skills and MCP servers. Usage: /gobby <skill> [args] or /gobby mcp <server> [args]

Manage subagent spawning - spawn, cancel, list, and check results of autonomous agents.

Quickly create a bug task. Usage: /bug <title> [description]

Quickly create a chore/maintenance task. Usage: /chore <title> [description]

Quick reference when blocked by "no active task" error. Shows how to create or claim a task before using Edit, Write, or NotebookEdit tools.

This skill should be used when the user asks to "/gobby clones", "create clone", "spawn in clone". Manage git clones for isolated parallel development - create, spawn agents, sync, merge, and delete.

Commit message format and task close workflow. Use when ready to commit or closing tasks.

Progressive disclosure pattern for MCP tools and skills. Use when tools aren't found or you need to discover available capabilities.

This skill should be used when the user asks to "/gobby doctor", "run doctor", "security audit", "health check", "systems check", "run diagnostics". Run comprehensive systems check and security audit verifying CLI and MCP tools work correctly.

Quickly create an epic (parent task for large features). Usage: /epic <title> [description]

This skill should be used when the user asks to \"evaluate agent performance\", \"build test framework\", \"measure agent quality\", \"create evaluation rubrics\", or mentions LLM-as-judge, multi-dimensional evaluation, agent testing, or quality gates for agent pipelines.

Use when user asks to '/gobby expand', 'expand task', 'break down task', 'decompose task'. Expand a task into subtasks using codebase analysis and visible LLM reasoning. Survives session compaction.