robobuilder

Plugin robobuilder from marketplace robo-coop-tools in Robo-Co-op/robobuilder-standard. Commands read from the repo's manifest — yesterday.

Skills (42)

• install-companions

  [Meta-4] Install a curated set of Anthropic official skill packs alongside robobuilder, by role. Presets defined in company.yaml. Use when user invokes /robobuilder:install-companions or asks \"what skills should I install on top of robobuilder?\".

• start

  [Meta-1] Onboarding wizard for first-time robobuilder users. Asks about role, experience level, and current task, then suggests the right starting skills. Use this when a user invokes /robobuilder:start, when they ask \"how do I use this plugin?\", or when they seem lost in the skill catalog.

• tune-claude-md

  [Meta-2] Personalize the user's ~/.claude/CLAUDE.md by intelligently merging it with the Robo Co-op best-practice baseline. Classifies each baseline section as UPGRADE / KEEP / ADD, prompts for confirmation, backs up the original, and applies. Use when user invokes /robobuilder:tune-claude-md, after

• upgrade

  [Meta-3] Semver-aware upgrade of robobuilder. Pulls latest from the official repo, shows a diff preview of skills/agents/hooks that changed, prompts for approval per change, then applies. Re-runs tune-claude-md if baseline updated. Use when user invokes /robobuilder:upgrade or when they ask \"is rob

• setup

  [P0-1 Foundation] Sets up an `## Agent skills` block in AGENTS.md/CLAUDE.md and `docs/agents/` so the engineering skills know this repo's issue tracker (GitHub or local markdown), triage label vocabulary, and domain doc layout. Run before first use of `to-issues`, `to-prd`, `triage`, `diagnose`, `td

• ubiquitous-language

  [P0-2 Foundation] Extract a DDD-style ubiquitous language glossary from the current conversation, flagging ambiguities and proposing canonical terms. Saves to UBIQUITOUS_LANGUAGE.md. Use when user wants to define domain terms, build a glossary, harden terminology, create a ubiquitous language, or me

• diagnose

  [P0.5-2 Investigate] Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says \"diagnose this\" / \"debug this\", reports a bug, says something is broken/throwing/failing, or describes a performa

• health

  [P0.5-3 Investigate] Code quality dashboard. Wraps existing project tools (type checker, linter, test runner, dead code detector, shell linter), computes a weighted composite 0-10 score, and tracks trends over time. Use when: "health check", "code quality", "how healthy is the codebase", "run all ch

• zoom-out

  [P0.5-1 Investigate] Tell the agent to zoom out and give broader context or a higher-level perspective. Use when you're unfamiliar with a section of code or need to understand how it fits into the bigger picture.

• design-an-interface

  [P1-3 Design] Generate multiple radically different interface designs for a module using parallel sub-agents. Use when user wants to design an API, explore interface options, compare module shapes, or mentions \"design it twice\".

• grill-me

  [P1-1 Design] Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions \"grill me\".

• grill-with-docs

  [P1-2 Design] Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.

• plan-eng-review

  [P1-4 Design] Eng manager-mode plan review. Lock in the execution plan — architecture, data flow, diagrams, edge cases, test coverage, performance. Walks through issues interactively with opinionated recommendations. Use when asked to "review the architecture", "engineering review", or "lock in the

• to-issues

  [P1-6 Design] Break a plan, spec, or PRD into independently-grabbable issues on the project issue tracker using tracer-bullet vertical slices. Use when user wants to convert a plan into issues, create implementation tickets, or break down work into issues.

• to-prd

  [P1-5 Design] Turn the current conversation context into a PRD and publish it to the project issue tracker. Use when user wants to create a PRD from the current context.

• prototype

  [P2-1 Prototype] Build a throwaway prototype to flush out a design before committing to it. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use when the user wants to prototype, sanity-c

• browse

  [P3-4 Implement] Fast headless browser for QA testing and site dogfooding. Navigate any URL, interact with elements, verify page state, diff before/after actions, take annotated screenshots, check responsive layouts, test forms and uploads, handle dialogs, and assert element states. ~100ms per comma

• caveman

  [P3-3 Implement] Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman.

• learn

  [P3-5 Implement] Manage project learnings. Review, search, prune, and export what RoboBuilder has learned across sessions. Use when asked to "what have we learned", "show learnings", "prune stale learnings", or "export learnings". Proactively suggest when the user asks about past patterns or wonders

• tdd

  [P3-2 Implement] Test-driven development with red-green-refactor loop. Use when user wants to build features or fix bugs using TDD, mentions \"red-green-refactor\", wants integration tests, or asks for test-first development.

• triage

  [P3-1 Implement] Triage issues through a state machine driven by triage roles. Use when user wants to create an issue, triage issues, review incoming bugs or feature requests, prepare issues for an AFK agent, or manage issue workflow.

• improve-codebase-architecture

  [P3.5-1 Refactor] Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-na

• request-refactor-plan

  [P3.5-2 Refactor] Create a detailed refactor plan with tiny commits via user interview, then file it on the project's configured issue tracker. Use when user wants to plan a refactor, create a refactoring RFC, or break a refactor into safe incremental steps.

• cross-review

  [P4-2 Review] AI cross-review — multi-round parallel review with 4 perspectives that keeps iterating until zero findings (0 critical + 0 medium for 2 consecutive rounds; stop when rounds stop surfacing new defects, typically around five). Expensive; use only before important merges

• cso

  [P4-4 Review] Chief Security Officer mode. Infrastructure-first security audit: secrets archaeology, dependency supply chain, CI/CD pipeline security, LLM/AI security, skill supply chain scanning, plus OWASP Top 10, STRIDE threat modeling, and active verification. Two modes: daily (zero-noise, 8/10

• diff-review

  [P4-1 Review] Daily-driver diff review. Reviews the current branch's diff with 3 parallel subagents (code-simplifier / test-writer / security-auditor) and merges their findings into one prioritized fix list

• grill

  [P4-3 Review] Adversarial review of the latest changes (diff). Interrogates the code mercilessly to expose oversights, false assumptions, and vulnerabilities — lists 5+ failure modes and ends with a SHIP / PATCH FIRST / REWRITE verdict

• canary

  [P5-3 Ship] Post-deploy canary monitoring. Watches the live app for console errors, performance regressions, and page failures using the browse daemon. Takes periodic screenshots, compares against pre-deploy baselines, and alerts on anomalies. Use when: "monitor deploy", "canary", "post-deploy check

• handoff

  [P5-4 Ship] Compact the current conversation into a handoff document for another agent to pick up.

• land-and-deploy

  [P5-2 Ship] Land and deploy workflow. Merges the PR, waits for CI and deploy, verifies production health via canary checks. Takes over after /ship creates the PR. Use when: "merge", "land", "deploy", "merge and verify", "land it", "ship it to production". (robobuilder-local) - Bash - Read - Write -

• secret-transfer

  [P5-6 Ship] Deliver API keys, tokens and passwords from GitHub secrets to a server's .env without ever exposing them. Use when: \"add the key to the server\", \"update .env\", \"deploy secrets\", \"rotate a token\".

• ship

  [P5-1 Ship] Ship workflow: detect + merge base branch, run tests, review diff, bump the version, update CHANGELOG, commit, push, create PR. Use when asked to "ship", "deploy", "push to main", "create a PR", "merge and push", or "get it deployed". Proactively invoke this skill (do NOT push/PR directl

• write-a-skill

  [P5-5 Ship] Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.

• playbook-bug-fix

  [PB-2] Orchestrates a clean bug-fix workflow by invoking robobuilder skills in sequence. Steps diagnose → tdd → diff-review → ship. Use when user reports a bug and wants the guided fix flow rather than piecing it together.

• playbook-new-feature

  [PB-1] Orchestrates the full new-feature workflow by invoking robobuilder skills in sequence via the Skill tool. Steps grill-me → to-prd → to-issues → tdd (parallel via Agent teams) → diff-review → cross-review → grill → ship. Use this when a user wants to build a new feature end-to-end and asks for

• playbook-review-deep

  [PB-3] Three-tier escalating review for critical merges. Runs diff-review → cross-review → grill in sequence, with user confirmation between tiers. Use before merging high-impact branches (security, payments, auth, multi-system migrations).

• blueprint-sync

  [Util-2] Detects drift between the codebase and its design documents (PRD, architecture docs, ADRs, README), then updates or creates docs to reflect reality. Use when: code has drifted from the blueprint, after landing a big feature, running a retrospective after ship, or maintaining living document

• btw

  [Util-5] Lightweight one-shot question. Answers immediately with minimal context — no heavy reviews, no investigations, no subagents

• context-restore

  [Util-4] Restore working context saved earlier by /context-save. Loads the most recent saved state (across all branches by default) so you can pick up where you left off — even across Conductor workspace handoffs. Use when asked to "resume", "restore context", "where was I", or "pick up where I left

• context-save

  [Util-3] Save working context. Captures git state, decisions made, and remaining work so any future session can pick up without losing a beat. Use when asked to "save progress", "save state", "context save", or "save my work". Pair with /context-restore to resume later. Formerly /checkpoint — rename

• export

  [Util-6] Concatenate the entire repository's source code into a single file, for uploading to claude.ai Deep Research or sharing externally

• guard

  [Util-1] Full safety mode: destructive command warnings + directory-scoped edits. Combines /careful (warns before rm -rf, DROP TABLE, force-push, etc.) with /freeze (blocks edits outside a specified directory). Use for maximum safety when touching prod or debugging live systems. Use when asked to "g

Subagents (9)

• code-simplifier

  Reviews a recent diff for reuse, redundancy, over-abstraction, excessive guards, and deep nesting. Returns concrete before/after diff suggestions. Use this immediately after Edit/Write operations, when the user asks "is this too complex?", or as part of /diff-review.

• codebase-explorer

  Maps unfamiliar or large codebases — module relationships, call graphs, key abstractions, and entry points. Use at the start of work on an existing repo, or when an existing repo grew enough that the mental model is stale. Extends the built-in Explore agent with structured reporting for the Phase 0.

• design-critic

  Adversarial design reviewer. Given a design document, PRD, or proposal, attacks it from multiple angles to surface assumptions, missing requirements, and failure modes. Complements the interactive /grill-me by running offline as a sub-agent. Use before locking a design, before architectural commits,

• e2e-tester

  Designs, generates, and runs Playwright end-to-end tests. Use this when adding a critical user flow, before deploys, or when regression coverage is thin. Always run as a subagent so the massive execution logs don't pollute the main context.

• release-notes-writer

  Generates CHANGELOG entries and PR descriptions from a branch's commit history. Groups commits by type (feat / fix / refactor / docs / chore), expands short messages with context from the diff, and produces a user-facing summary. Use during /ship or /land-and-deploy, or when preparing a release PR.

• requirements-validator

  Reviews a PRD or requirements document for gaps, contradictions, ambiguity, and untestability. Use after /to-prd is generated, before /to-issues is invoked. Catches requirements problems before they become implementation problems.

• security-auditor

  Reviews the current branch diff (or specified files) against OWASP Top 10 2021. Use this immediately after changes to route handlers, auth/authz, DB queries, or anything handling external input. Auto-invoke when a diff touches /api, /auth, /admin, or DB schema files.

• tdd-pair

  Enforces TDD discipline during implementation. Watches for "wrote code without a failing test first" and stops the user. Use whenever /tdd is invoked, or when the user is implementing a feature and you suspect they're cutting corners on tests.

• test-writer

  Audits test coverage of changed files, flags missing cases, and generates the high-priority tests. Use this right after a new feature lands or a bug fix is applied — before /diff-review.

Hooks (6)

• SessionStart

  python "${CLAUDE_PLUGIN_ROOT}/scripts/notification.py" session-start

• PreToolUse

  python "${CLAUDE_PLUGIN_ROOT}/scripts/block_secrets.py"

• PostToolUse

  python "${CLAUDE_PLUGIN_ROOT}/scripts/auto_format.py"

• Notification

  python "${CLAUDE_PLUGIN_ROOT}/scripts/notification.py"

• PreCompact

  python "${CLAUDE_PLUGIN_ROOT}/scripts/memory_consolidate.py" pre-compact

• SessionEnd

  python "${CLAUDE_PLUGIN_ROOT}/scripts/memory_consolidate.py" session-end

Similar plugins