By Acendas
Turn your idea into working code — Shipyard runs Claude Code through a spec-driven agile sprint lifecycle: discuss features, plan sprints, build with TDD, review, and ship.
Executes sprint tasks by writing code with strict test-driven development (write tests first, then code). Follows acceptance criteria from task specs.
Adversarial reviewer that challenges feature specs and sprint plans before user approval. Multi-persona critique with structured findings. Read-only — never modifies artifacts.
Discovery analyst for /ship-discuss Phase 1.5b. Loads challenge / edge-case / NFR / failure-mode methodology references and the feature draft, returns a structured findings list. Read-only — never modifies artifacts.
Deep-dive investigation agent. Receives ONE code review finding and confirms or refutes it with evidence. Reads call sites, traces data flow, checks tests. Returns a verdict.
Investigates technical unknowns, codebase patterns, and external APIs. Read-only with respect to the codebase — searches and reports but never modifies code. Write tool is scoped by contract to findings docs under <SHIPYARD_DATA>/research/ only, never anywhere else. Dispatched by /ship-execute for kind: research sprint tasks (task-driven mode) and by /ship-discuss or /ship-sprint for inline technical investigation (free-form mode). Fire when a task has kind: research, or when a design decision needs tradeoff analysis, external API evaluation, or a codebase pattern scan before a sprint is planned.
View, groom, and manage the prioritized backlog sorted by RICE score. Use when the user wants to see the backlog, reprioritize features, run a grooming session, clean up stale items, declare backlog bankruptcy, or decide what to work on next.
Report a bug or production issue with minimal ceremony. Creates a spec entry and optional fix task. Use when the user reports something broken, a defect, unexpected behavior, a regression, or needs to file a hotfix for production. Also use for --hotfix emergency production issues.
Systematic debugging with persistent state that survives session breaks and /clear. Use when the user reports a bug, something isn't working, tests are failing, they're stuck on an error, or they want to investigate unexpected behavior. Also use when the user says 'debug', 'investigate', 'why is this broken', or 'help me fix this'.
Feature discovery — from quick idea capture to full spec with acceptance criteria. Use when the user mentions a new feature, a 'what if', a 'we should also', wants to discuss requirements, brainstorm, refine an existing feature, explore what to build next, define acceptance criteria, or jot down something for later.
Execute the current sprint by running tasks in waves with strict test-driven development (write tests first, then code). Supports solo, subagent, and team execution modes. Use when the user wants to start building, execute sprint tasks, run a specific task, apply a hotfix, or resume execution after a break.
Matches all tools
Hooks run on every tool call, not just specific ones
Executes bash commands
Hook triggers when Bash tool is used
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Modifies files
Hook triggers on file write and edit operations
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
Uses power tools
Uses Bash, Write, or Edit tools
_T_
______|░|___
_T_ / |░| \
__|░|_________ |░|░| |░| |
|░| \ |░|░| |░| ██████╗
|░| | |░| ██╔══██╗
|░| & |░| ██║ ██║
|░| ███████╗██╗ ██╗██╗██████╗ ░░╗ ░░╗ █████╗ ██████╗ |░| ██║ ██║
|░| ██╔════╝██║ ██║██║██╔══██╗╚░░╗ ░░╔╝ ██╔══██╗██╔══██╗ |░| ██████╔╝
|░| ███████╗███████║██║██████╔╝ ╚░░░░╔╝ ███████║██████╔╝ |░| ╚═════╝
|░| ╚════██║██╔══██║██║██╔═══╝ ╚░░╔╝ ██╔══██║██╔══██╗ |░|
|░| ███████║██║ ██║██║██║ ░░║ ██║ ██║██║ ██║ |░|
|░| ╚══════╝╚═╝ ╚═╝╚═╝╚═╝ ╚═╝ ╚═╝ ╚═╝╚═╝ ╚═╝ |░|
░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
You're copy-pasting requirements into chat windows. You're re-explaining context every session. You're manually checking if the AI actually built what you asked for. You're debugging code that passed the AI's own "tests." You're losing work when sessions crash. You're starting from scratch every Monday.
That's not AI-assisted development. That's you being the project manager for a junior dev with amnesia.
Shipyard is a full engineering org — planner, builders, reviewers, critics — that runs inside Claude Code. You describe what you want. Shipyard argues about the best approach, writes a spec, plans the sprint, builds everything test-first with parallel agents, then has a separate agent verify the work against the spec before you even see it.
You talk. Shipyard plans. Claude builds. You approve.
┌───────┐ ┌───────┐ ┌───────┐ ┌───────┐ ┌───────┐
│ IDEA │───►│ SPEC │───►│ PLAN │───►│ BUILD │───►│ SHIP │
│ │ │ │ │ │ │ │ │ │
│ you │ │ you + │ │ you + │ │ auto │ │ you │
│ talk │ │ claude│ │ claude│ │ │ │approve│
└───────┘ └───────┘ └───────┘ └───────┘ └───────┘
/discuss /discuss /sprint /execute /review
Feature specs, backlog grooming, sprint planning, test-driven execution, code review, retros, and releases — all through /ship-* slash commands. No browser tabs. No context switching. Just you and Claude building software.
Without Shipyard
|
With Shipyard
|
Every AI coding tool gives you a smart agent. Shipyard gives you a team that argues.
Before any plan reaches you, an adversarial critic runs a pre-mortem — imagining how this feature fails spectacularly in 3 months, extracting hidden assumptions, and challenging every design decision. Before any code ships, a separate reviewer verifies it against the spec. Before any test passes, mutation testing confirms the tests actually catch bugs — not just that they run green.
npx claudepluginhub acendas/shipyard --plugin shipyardAI-agent-driven Android debugger — exposes JDI/JDWP and a JVMTI agent as MCP tools for Claude Code to drive. v1.8 adds standalone, attach-free static analysis backed by SootUp 2.0.0: static_class_hierarchy / static_call_graph / static_cfg / static_package_graph return agent-facing nodes/edges/truncated/warnings plus ASCII and Mermaid renderings, with BFS depth/node-cap bounds, virtual-dispatch-target capping, and Kotlin-synthetic collapse (companions, lambdas, DefaultImpls, coroutine continuations). New /android-debugger:ad-graph skill resolves class_dirs across Gradle modules and android_api_level, with a confirm-then-write Mermaid-persistence flow — routed as a 'Zeroth check' for pure structural questions ('what implements X', 'call graph of Y', 'class hierarchy of Z') ahead of the Debug-Plan shapes, and available mid-plan as a read-only side-query. v1.7 shipped Debug Plans: the agent authors a Plan declaring setup breakpoints, on_event handlers with actions (snapshot, FEEL expressions, eval_method, resume/step, yield_when, abort_when, log, set_var), hypotheses graded matched/contradicted/inconclusive, and a streaming plan_progress event feed. New MCP tools: run_debug_plan, pause_plan, abort_plan, validate_plan, list/save/load/delete_plan. New dbg.* FEEL namespace (instance_count, is_reachable, thread_state, frame_count, has_capability, elapsed_ms, logcat_since). Hybrid concurrency: read-only inspection passes through during a plan; mutating tools return vm_in_plan. Recipe skills /ad-catch /ad-trace /ad-walk /ad-bisect-flaky author plans; /ad-explain stays imperative; /ad-investigate dispatches templates and handles yield→abort→re-author. Plus everything from v1.0–v1.6: JVMTI deepening (native-speed heap walks count_instances / iterate_heap_by_class / find_referrers / find_referrer_chain auto-routing, line-rate method tracing with leaky-bucket throttling + arg + return-value capture, class-allowlisted allocation tracing, agent v3 JSON-RPC), HotSwap (hot_swap_class + hot_swap_classes + hot_swap_revert), the goal-driven /android-debugger:patch skill, breakpoints (line, conditional, hit-count, logpoint, exception, method, field, class-load), watches, step, FEEL expression evaluation (kfeel), eval_method with mutation refusal, frame snapshots, logcat, heap dump, session persistence, and the C++ JVMTI agent (3 ABIs). Cross-platform: macOS, Linux, Windows.
Acendas Atlassian Suite — unified Jira Cloud, Confluence Cloud, Bitbucket Cloud, and QMetry Test Management integration for Claude Code. Jira + Confluence + Bitbucket + QMetry tools, workflow skills, and a multi-scanner PR review pipeline.
An agent-routed harness for end-to-end software product development
Spec-driven development for big features. When features get too big, plan mode gets too vague—leading to hallucinations during implementation. ShipSpec replaces vague plans with structured PRDs, technical designs, and ordered tasks that keep Claude grounded.
Helder's personal SDLC toolbelt for AI coding agents — from PRD to ship. Bundles the tracer-bullet workflow alongside TDD, code review, audits, and shipping skills.
Persona-driven AI development team: orchestrator, team agents, review agents, skills, slash commands, and advisory hooks for Claude Code
Autonomous multi-agent development framework with spec-driven sprints and convergent iteration
Give soul to your workflow. 58 AI-powered skills across 17 roles — PM, Dev, Backend, Frontend, QA, UX, Data, Detect, WordPress, Release, Security, DevOps, and Core. Spec-to-ship pipeline: scaffold, implement, test, secure, deploy. Features two-phase workflow with human approval, quality-reviewer agent, token optimization, and continuous improvement via LEARN.md system.