escapement

Plugin escapement from marketplace escapement in alexander-vyh/escapement. Commands read from the repo's manifest — 5 days ago.

Skills (11)

• beads-execution

  Use only when the user explicitly asks to execute, work on, run, or start tracked Beads work. Informational mentions and task IDs alone do not authorize execution.

• beads-worktree

  Use when creating or completing an isolated worktree, or checking Beads task-state resolution inside one. Escapement owns creation policy; Beads remains tracker state.

• behavioral-test-oracle-review

  Use when implementing, writing tests, modifying tests, reviewing tests, fixing bugs, or changing business logic, reports, data models, APIs, auth, UI flows, jobs, or integrations where tests must prove user or business behavior rather than implementation details.

• brainstorming

  Enhanced brainstorming — challenges whether work should be done at all, rotates creative lenses to fight semantic clustering, and routes to /discovery or work-breakdown based on complexity.

• discovery

  Pre-design-doc discovery skill. Fills the gap between "we should build X" and "here is the design doc." Produces genuine thinking through adversarial questions, not checkbox compliance. Requires a confirmed problem framing as input (Input Gate) — then drafts solutions via a draft-and-react model. Us

• dispatching-parallel-agents

  Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies

• gate-design

  Use when adding or modifying a hook, gate, denial/permission message, or waiver/exemption/override path; when deciding whether to keep, revise, or retire a rule or gate; or when auditing a gate or triaging a complaint that a gate causes friction. The implementation manual for building gates that are

• huh

  Type "huh" (or "huh?") the moment a message from the agent doesn't land. The agent re-pitches what it just said in plain language, using this repo's own terms from docs/VOCABULARY.md instead of jargon. User-invoked only — never fire this on your own initiative.

• subagent-driven-development

  Use when executing implementation plans with independent tasks in the current session

• vocab

  Use whenever you set out to understand or research an UNFAMILIAR external domain — building something is NOT required (also preparing for a decision or meeting, forming a position, an investigation, or designing) — when getting the field's framing right matters and a load-bearing distinction hinges

• work-breakdown

  Translate a validated design (from openspec/changes/ or docs/plans/) into a beads task graph with outcome-based acceptance criteria, failure modes, scope boundaries, and spec traceability via --spec-id. Invoked after the walking skeleton validates assumptions.

Slash commands (8)

• /escapement:brainstorm

  Enhanced brainstorming — challenges whether work should be done, rotates creative lenses, routes to /discovery or writing-plans

• /escapement:discovery

  Run discovery to create a design doc with adversarial planning — problem, non-goals, riskiest assumption, walking skeleton

• /escapement:opsx:apply

  Implement tasks from an OpenSpec change (Experimental)

• /escapement:opsx:archive

  Archive a completed change in the experimental workflow

• /escapement:opsx:explore

  Enter explore mode - think through ideas, investigate problems, clarify requirements

• /escapement:opsx:propose

  Propose a new change - create it and generate all artifacts in one step

• /escapement:review

  Dispatch parallel review agents for code review

• /escapement:work-breakdown

  Translate a design doc into a beads task graph with outcome-based acceptance criteria, failure modes, and ambiguity flags

Subagents (2)

• adversarial-reviewer

  Hostile, deeply expert code and design reviewer. Runs as isolated subagent with no conversation history — only sees the artifact to review. Personally motivated to find failures.

• test-quality-reviewer

  Reviews test files for assertion quality. Flags tests that only verify structure (not NULL, count > 0) without asserting specific expected business outcomes. Use after implementation to audit test quality before PR.

MCP servers (1)

• serena

  runs: uvx --from serena-agent==1.7.0 serena start-mcp-server --project-from-cwd --enable-web-dashboard false --open-web-dashboard false --context claude-code

Hooks (30)

• PostToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/design_doc_location_guard.py"

• PostToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/design_doc_location_guard.py"

• PostToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/shadow_verifier.py"

• PostToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/shadow_verifier.py"

• PostToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/shell_write_gate.py"

• PostToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/test_reminder.py"

• PostToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/test_reminder.py"

• PostToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/harness/bin/schedule_wakeup_bridge.py"

• PostToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/harness/bin/task_mode_entry.py"

• PostToolUseFailure

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/shell_write_gate.py"

• PreCompact

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/escapement_session_context.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/test_oracle_brief_gate.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/review_gate.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/beads_worktree_guard.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/root_checkout_guard.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/root_checkout_guard.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/root_checkout_guard.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/root_checkout_guard.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/cache_write_guard.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/context_burn_detector.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/context_burn_detector.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/context_burn_detector.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/context_burn_detector.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/discovery-close-gate.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/discovery-gate.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/discovery_input_gate.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/discovery_input_gate.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/enforce_named_agents.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/epic_coverage_gate.py"

• PreToolUse

  python3 -B "${CLAUDE_PLUGIN_ROOT}/hooks/gate_design_nudge.py"

Similar plugins