cast

Plugin cast from marketplace cast in ek33450505/claude-agent-team. Commands read from the repo's manifest — 4 days ago.

Skills (19)

• agent-registry

  Full CAST agent registry with model assignments. Load when building an Agent Dispatch Manifest, orchestrating multi-agent workflows, or when you need to verify which model a specific agent runs on.

• briefing-writer

  Assemble morning briefing sections into a structured markdown file. Use when writing the final briefing output after all data sources have been gathered.

• careful-mode

  Activate careful mode — require explicit confirmation before any Write, Edit, or Bash operations. Use when working on production code or sensitive systems.

• cast-audit

  Comprehensive CAST codebase audit — bugs, security, performance, test coverage. Dispatches 4 parallel researchers to scan the codebase and collates findings into a dated report. Manual invocation only; monthly scheduling is a design intention (helper not yet shipped).

• cast-conventions

  Shared CAST conventions for all agents. Loaded automatically via agent frontmatter.

• deep-research

  Deep research harness — fan-out web searches, fetch sources, adversarially verify claims, synthesize a cited report. When the user wants a deep, multi-source, fact-checked research report on any topic. BEFORE invoking, check if the question is specific enough to research directly — if underspecified

• freeze-mode

  Freeze mode — read-only session, no file modifications allowed. Use for exploration, research, or review-only sessions.

• git-activity

  Scan project repositories for yesterday's git activity. Use when summarizing development progress for briefings, reports, standups, or when user asks "what did I work on yesterday" or "recent commits".

• laconic

  Terse output mode. Compresses Claude prose to minimum tokens while preserving technical accuracy. Three intensity levels: lite, full (default), ultra.

• merge

  Git merge, rebase, and conflict resolution specialist. Use when merging feature branches, rebasing onto main, resolving merge conflicts, merging PRs via gh CLI, or cleaning up merged worktrees and branches. Hard-blocks force-merges to main/master without explicit approval.

• orchestrate

  Execute a CAST plan by reading the Agent Dispatch Manifest and dispatching agents in dependency order — parallel batches simultaneously, sequential batches one at a time. Pass a plan file path, 'next' for the most recent plan, or 'resume' to continue from a checkpoint.

• plan

  Activate the heavy planner→/orchestrate chain — write a structured plan file with Agent Dispatch Manifest, then invoke the /orchestrate skill to execute it in waves. Reserve for genuinely multi-file / multi-hour / multi-agent work (large refactors, migrations, features spanning many files). For sing

• python-conventions

  Python and shell-script conventions for CAST projects. Load when writing, reviewing, or debugging Python scripts, hook scripts, or files under scripts/ or bin/. Covers stdlib-only policy, DB access patterns, exit codes, idempotency, stdin parsing, and hook script conventions.

• seo-checklist

  SEO and GEO (Generative Engine Optimization) checklist for web projects. Load when reviewing page metadata, structured data, technical SEO, Core Web Vitals, and generative-engine optimization. Covers classic search visibility and LLM-crawler optimization.

• seo-index-census

  Index-coverage census discipline for sites with many programmatic routes — when to run a census, how to read discovered-not-indexed vs crawled-not-indexed, GSC URL Inspection quota arithmetic, stratified sampling, drift baselines, and the pre-registered soak window. Load when investigating unindexed

• ship

  CAST ship skill

• stack-reference

  Full tech stack reference and dev command cheatsheet. Load when you need to look up exact package names, commands, or project paths.

• typescript-conventions

  TypeScript, React, and UI accessibility conventions for CAST projects. Load when writing, reviewing, or debugging TypeScript/TSX/React code, or when checking accessibility (a11y, aria, focus, contrast, keyboard navigation). Covers file naming, type patterns, hooks, component conventions, testing app

• wizard

  Multi-step workflow with human-approval gates before destructive operations. Use when a task involves irreversible actions like mass file writes, database migrations, git force-push, dependency removal, or infrastructure changes.

Slash commands (19)

• /cast:agents

  CAST agents command

• /cast:bash

  CAST bash command

• /cast:cast

  CAST cast command

• /cast:ci-watch

  CAST ci-watch command

• /cast:commit

  CAST commit command

• /cast:debug

  CAST debug command

• /cast:devops

  CAST devops command

• /cast:docs

  CAST docs command

• /cast:doctor

  CAST doctor command

• /cast:feature

  CAST feature command

• /cast:laconic

  Toggle laconic terse-output mode. Usage: /laconic [lite|full|ultra|off]

• /cast:merge

  CAST merge command

• /cast:orchestrate

  CAST orchestrate command

• /cast:plan

  CAST plan command

• /cast:research

  CAST research command

• /cast:review

  CAST review command

• /cast:roadmap

  CAST roadmap command

• /cast:secure

  CAST secure command

• /cast:test

  CAST test command

Subagents (22)

• backend-writer

  Implementation specialist for backend feature work, bug fixes, and planned changes. Receives tasks from planner or orchestrating session, writes production code following project conventions, mandatorily chains code-reviewer after each logical unit, writes tests inline, and dispatches the commit age

• bash-specialist

  Shell scripting specialist for CAST hook scripts, BATS tests, and automation scripts. Use when writing new hook scripts, BATS test suites, reviewing shell code for correctness, debugging hook behavior, or extending CAST automation. Knows CAST-specific conventions: exit codes, escape hatches, hookSpe

• code-reviewer

  Per-unit code-quality and security review of a single logical change mid-flight, any language. Use immediately after each backend-writer, frontend-writer, or debugger unit, before commit. Not for full-PR review (use pr-reviewer) and not for deep React/TypeScript or accessibility review (use frontend

• commit

  Git commit specialist. Use after completing a feature, fix, or meaningful change. Reads staged changes, writes a semantic commit message, and commits cleanly.

• db-reader

  Read-only data-analysis specialist. Use for: SQL queries, data exploration, and reporting against BigQuery or SQLite. Restores the former db-reader role.

• debugger

  Root-cause debugging of concrete failures — a reproduced error, a stack trace, a failing test, or observably wrong runtime behavior. Use when investigation needs more than one inline tool call. NOT for code-quality concerns (use code-reviewer), writing tests for passing code (use test-writer), or re

• devops

  CI/CD pipeline management and GitHub Actions workflow authoring.

• docs

  Documentation specialist. Handles README audits/rewrites and doc updates after code changes. Absorbs the former readme-writer and doc-updater roles.

• email-drafter

  Email and portfolio-sync specialist. Drafts emails from bullet points, context, or instructions (compose, reply, gmail) — creates Gmail drafts for review, never sends. Also syncs showcase repo READMEs with actual project state (readme stats, showcase, portfolio update). Recovers the former email-dra

• eval-writer

  Eval and benchmark fixture author for Claude API and CAST agent prompts. Use proactively when an agent definition, system prompt, or routing rule changes. Generates YAML eval cases with runnable graders that catch prompt-level behavior drift.

• frontend-qa

  Deep React 19 and TypeScript review of .tsx/.ts files in dashboard projects only — prop typing, TanStack Query hook usage, frontend-to-backend API-contract alignment, Vitest coverage gaps, and a11y basics. Dispatch manually or via a planner manifest for .tsx/.ts review. Does NOT cover style/naming/f

• frontend-writer

  Implementation specialist for frontend feature work, bug fixes, and planned changes. Receives tasks from planner or orchestrating session, writes production code following project conventions, mandatorily chains code-reviewer after each logical unit, writes tests inline, and dispatches the commit ag

• infra-writer

  Docker/containerization, infrastructure-as-code (Terraform, CloudFormation stubs), deployment configuration, and environment management.

• merge

  PR lifecycle agent. When chained from push: watches CI checks on the current PR and stops for confirmation before squash-merge. When invoked directly: handles local branch merges, rebases, conflict resolution, and worktree cleanup. Hard-blocks force-merges to main/master without explicit approval.

• migration-reviewer

  Database schema change reviewer. Analyzes migration files for safety, generates rollback plans, validates ordering, and checks for data-loss risks. Works with SQLite, PostgreSQL, and MS SQL Server. Read-only reviewer.

• planner

  Planning specialist that converts feature requests into specs and ordered task breakdowns. Reserve for genuinely multi-file / multi-hour / multi-agent work (large refactors, migrations, features spanning many files). For single-session-sized work (one or a few files), use built-in plan mode (shift-t

• pr-reviewer

  Whole-PR review at PR-open time only — reads the full multi-commit diff, commit-message coherence, scope creep, coverage gaps, and breaking-change surface. Dispatch only after `gh pr create` or an explicit end-to-end PR review request. Not for mid-flight single-unit review (use code-reviewer) or Rea

• report-writer

  Status/chain-reporting specialist. Generates weekly status updates, project health checks, and multi-agent chain execution summaries. Split off from the former docs agent.

• researcher

  Multi-purpose research specialist. Use for: codebase exploration, web research, and technology comparisons. Absorbs the former explore and data-scientist roles.

• security

  Security review specialist. Use when writing code that handles user input, auth, API keys, database queries, or external data. Scans for OWASP top 10, secrets exposure, and stack-specific vulnerabilities.

• test-runner

  Test execution gate. Runs the project test suite and gates the chain on real exit codes. Dispatched by the orchestrator before commit. Does NOT write tests — use backend-writer/frontend-writer for that. On failure, reports failing test names and exit code; the orchestrator dispatches debugger when n

• test-writer

  Test design specialist. Writes test suites for existing code — happy path, edge cases, and error states. Detects the project's test framework and follows existing conventions. Use after backend-writer or frontend-writer completes a logical unit.

Hooks (27)

• SessionStart

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-plugin-bootstrap.sh"

• SessionStart

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-session-start-hook.sh"

• SessionStart

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-plan-resume-hook.sh"

• SessionStart

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-resume-inject-hook.sh"

• SessionStart

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-time-context-hook.sh"

• SessionStart

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-session-start-journal.sh"

• SessionStart

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-session-start-health.sh"

• UserPromptSubmit

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-user-prompt-hook.sh"

• UserPromptSubmit

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-time-drift-hook.sh"

• PostToolUseFailure

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-tool-failure-hook.sh"

• PostToolUse

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/post-tool-hook.sh"

• PostToolUse

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-audit-hook.sh" --mode post

• ConfigChange

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-config-change-guard.sh"

• PreToolUse

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-audit-hook.sh" --mode pre

• PreToolUse

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; python3 -E -s "${CLAUDE_PLUGIN_ROOT}/scripts/cast-pretool-dispatch.py"

• PreToolUse

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/write-guards.sh"

• PreCompact

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-precompact-guard.sh"

• PreCompact

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-precompact-memory-save.sh"

• Stop

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-journal-session-end.sh"

• SessionEnd

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-session-end.sh"

• SubagentStart

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-subagent-start-hook.sh"

• SubagentStop

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-subagent-stop-hook.sh"

• SubagentStop

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-subagent-worktree-check.sh"

• StopFailure

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-stop-failure-hook.sh"

• TaskCreated

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-task-created-hook.sh"

• TeammateIdle

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-teammate-idle-hook.sh"

• TaskCompleted

  if [ -f "$HOME/.claude/config/cast-hook-owner" ]; then exit 0; fi; bash "${CLAUDE_PLUGIN_ROOT}/scripts/cast-task-completed-hook.sh"

Similar plugins