Diagnose and fix persistent behavioral issues in Claude sessions using a CBT framework. Eleven tools with adaptive graduation, behavioral experiments, positive reinforcement, and relapse prevention.
npx claudepluginhub jerrod/agent-plugins --plugin therapistClaude Code plugin marketplace by jerrod. Tools for agent-operated software engineering — readiness scoring, code review, and behavioral correction.
Grades a repository against a 100-point rubric across 12 dimensions using deterministic bash checkers and judgment subagents. Produces a machine-readable JSON report and a human-readable markdown report per run.
See plugins/axr/README.md for details.
Runs four specialized reviewers — architect, security, correctness, style — against a pull request diff in parallel, then aggregates findings into a single verdict with deduplication and severity sorting.
⚠ Requires
CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1. revue depends on Claude Code's experimental agent-team feature to spawn its four reviewers concurrently. Without this env var set, thereview-prskill cannot dispatch subagents. Add the export to your shell profile so every session has it:export CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1
See plugins/revue/README.md for details.
A CBT-adapted intervention framework for sessions where Claude repeatedly violates explicit rules. Bundles a /therapist slash command, ambient hooks that catch rationalization phrases live at Write/Edit/Bash tool use, and a reference toolbox of eleven techniques.
See plugins/therapist/README.md for details.
Enforces the brainstorm → plan → pair-build → review → ship workflow with 24 skills, 23 subagents, 8 lifecycle hooks, and executable gates at every checkpoint. File size, coverage, complexity, lint, and test-quality are all script-verified. PR descriptions embed proof artifacts any reviewer can independently re-run. Hard fork of arqu-co/rq at v1.29.8.
See plugins/sdlc/README.md for details.
# Install the marketplace in Claude Code
# /plugin → Add Marketplace → jerrod/agent-plugins
# Validate the marketplace + every plugin
bin/validate
# Run linting (shellcheck + JSON + frontmatter)
bin/lint
# Run tests (validates all checkers produce schema-valid JSON)
bin/test
In Claude Code:
/plugin → Add Marketplace → jerrod/agent-pluginsexport CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 in your shell profileThe marketplace is structured as:
plugins/<name>/ — each plugin self-contained (manifest, commands/skills/agents, scripts, docs, README, CLAUDE.md)bin/ — marketplace-level gate scripts that validate every plugin.claude-plugin/marketplace.json — marketplace manifestPlugins may use any of commands/, skills/, or agents/ as their entry point. scripts/ is optional (agent-team plugins like revue have none).
See CLAUDE.md for workflow conventions. All changes go through the sdlc workflow: brainstorm → plan → pair-build → review → ship.
Battle-tested Claude Code plugin for engineering teams — 38 agents, 156 skills, 72 legacy command shims, production-ready hooks, and selective install workflows evolved through continuous real-world use
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
AI-supervised issue tracker for coding workflows. Manage tasks, discover work, and maintain context with simple CLI commands.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Claude + Google Stitch workflow toolkit with MCP integration (prompt authoring, screen generation, design extraction)
Context-Driven Development plugin that transforms Claude Code into a project management tool with structured workflow: Context → Spec & Plan → Implement