ac

Plugin ac from marketplace ac in anilcancakir/claude-code. Commands read from the repo's manifest — today.

Skills (19)

• agent-creator

  Create or audit a Claude Code subagent: tools, model and effort, permissions, memory, isolation, and the system-prompt body.

• architecture-md-writer

  Write or audit an ARCHITECTURE.md for agents: boundaries, invariants with verify commands, flows and a change guide.

• auto

  Away mode: carry the run in progress, or a new request, to completion unattended, ending at one verification gate.

• claude-md-rules-creator

  Write or audit CLAUDE.md, CLAUDE.local.md and .claude/rules files: scope, paths, imports, size, and why a rule is ignored.

• commit

  Split the changes into atomic commits, preflight and push them; also rebase, squash and trace history. Pass --no-push when the user only asked to commit.

• coordinate

  Turn a large request into a reviewed roadmap of bundles and drive them through parallel ac:lane worktrees.

• design-md-writer

  Write or audit a visual-identity DESIGN.md: design tokens plus rationale in the Google Labs format, for UI agents.

• execute

  Run an approved ac plan to completion, wave by wave, with cost-tiered workers, per-wave verification and a final review.

• flow

  Run a medium or large request on the main thread: parallel research, interview, build in slices, oracle review.

• frontend-design

  Build distinctive, accessible web and mobile UI from a committed design direction and token plan.

• github-cli

  Work with GitHub repositories, issues, PRs, releases and Actions runs through the gh CLI instead of web fetches.

• init-project

  Set up or audit a project's Claude Code files, project documents, permissions, lint hook and language server from one approved proposal.

• install

  Set up ac for your user, or upgrade an earlier setup: my-coding and my-language skills, the global CLAUDE.md and settings.json, each write gated.

• plan

  Plan a change: research in parallel, ask only the decisions the user must make, write a tiered plan. Writes no code.

• product

  Turn an idea into a feasibility-checked PRD and a parallel-ready roadmap through research and an interview.

• project-md-writer

  Write or audit a PROJECT.md (what, scope, constraints, decisions) for agents; commands and conventions go in CLAUDE.md.

• prompt-writer

  Write or audit a prompt another Claude will run: system prompts, subagent briefs, skill, command and agent bodies, CLAUDE.md and rules.

• qa

  Test recent work the way a QA engineer would: derive cases from the requirement and the diff, run them against the live app, fix critical and high findings, retest once.

• skill-creator

  Create or audit a Claude Code skill or slash workflow.

Subagents (13)

• auto-verifier

  Completion gate for one /ac:auto run: checks the frozen criteria and returns a verdict. Spawned only by /ac:auto; not for general tasks.

• design-director

  Read-only design director. Use this when a UI needs its visual direction chosen, its token plan committed, or its screenshots critiqued. Returns decisions, never code.

• explore

  Read-only code search agent for this repository. Use this when finding definitions, callers or usages needs more than about three queries or several naming conventions. Specify depth: quick, medium or thorough; add REUSE BIAS: to hunt for code to reuse. Returns file:line citations.

• lane

  Worktree-isolated lane that drives one roadmap bundle through ac:plan and ac:execute for the /ac:coordinate coordinator. Spawned only by /ac:coordinate; not for general tasks.

• librarian

  Read-only research agent for anything outside this repository: library behaviour, API contracts, framework idioms, OSS code, official docs. Specify depth: quick, medium or thorough. Returns URL citations with snippets; code in this repository is ac:explore.

• oracle

  Read-only reviewer and advisor. Use this when a plan, diff, report or config change needs checking against its sources, or a stalled decision needs settling. Returns ranked findings.

• plan-code-review

  Post-implementation review of an executed ac plan's diff; returns findings, no verdict. Spawned only by /ac:execute; not for general tasks.

• plan-reviewer

  Advisory review of one ac plan file before execution; returns findings, no verdict. Spawned only by /ac:plan; not for general tasks.

• plan-worker-junior-high

  Junior-high-tier ac plan worker: junior-shaped steps at borderline coupling or context depth. Spawned by /ac:execute and ac:flow; not for general tasks.

• plan-worker-junior

  Junior-tier ac plan worker, the default: one to three file changes in business logic or framework idiom. Spawned by /ac:execute and ac:flow; not for general tasks.

• plan-worker-quick

  Quick-tier ac plan worker: single-file mechanical steps. Spawned by /ac:execute and ac:flow; not for general tasks.

• plan-worker-senior

  Senior-tier ac plan worker: cross-layer changes, architecture, migrations and complex edge cases. Spawned by /ac:execute and ac:flow; not for general tasks.

• qa-tester

  Fresh-context QA engineer for one /ac:qa run: derives cases from the requirement and the diff, runs them against the live app, and reports findings with evidence. Spawned only by ac:qa; not for general tasks.

MCP servers (1)

• ac

  runs: node ${CLAUDE_PLUGIN_ROOT}/cli/ac.js mcp

Hooks (23)

• SessionStart

  "${CLAUDE_PLUGIN_ROOT}"/hooks/session-start-plan-state.sh

• SessionStart

  "${CLAUDE_PLUGIN_ROOT}"/hooks/session-start-working-rules.sh

• SessionStart

  "${CLAUDE_PLUGIN_ROOT}"/hooks/session-lease-sweep.sh

• Stop

  "${CLAUDE_PLUGIN_ROOT}"/hooks/stop-guard.sh

• Stop

  "${CLAUDE_PLUGIN_ROOT}"/hooks/stop-guard-auto.sh

• Stop

  "${CLAUDE_PLUGIN_ROOT}"/hooks/stop-guard-coordinate.sh

• Stop

  "${CLAUDE_PLUGIN_ROOT}"/hooks/stop-guard-announce.sh

• Stop

  "${CLAUDE_PLUGIN_ROOT}"/hooks/session-lease-sweep.sh

• SubagentStop

  "${CLAUDE_PLUGIN_ROOT}"/hooks/subagentstop-background-guard.sh

• SessionEnd

  "${CLAUDE_PLUGIN_ROOT}"/hooks/session-lease-sweep.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/pretooluse-file-scope.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/pretooluse-plan-mode-block.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/pretooluse-bash-guard.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/pretooluse-explore-budget.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/pretooluse-subagent-bash-guard.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/pretooluse-shell-syntax-guard.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/pretooluse-memory-file-gate.sh

• UserPromptSubmit

  "${CLAUDE_PLUGIN_ROOT}"/hooks/userpromptsubmit-session-title.sh

• PostToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/posttooluse-memory-index-budget.sh

• PostToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/posttooluse-cli-diagnostics.sh

• PostToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/posttooluse-cli-diagnostics.sh --async

• PostToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/posttooluse-memory-file-budget.sh

• PostToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/posttooluse-question-timeout.sh

Similar plugins