autonomous agents plugins

core0-io/nio

Plugin Claude Code

Nio — Execution assurance and observability for autonomous AI agents. Dynamic guard, static scan, OTEL collector.

GobbyAI/gobby

Plugin Claude Code

AST-aware code search, symbol navigation, and dependency graph analysis.

wrg32786/operator-kit

Plugin Claude Code

Small, focused Claude Code operator tools.

wrg32786/operator-kit

Plugin Claude Code

Five focused, least-privilege Claude Code subagents plus an optional keyword-triggered context loader.

sweatco/archie-hq

Plugin Claude Code

E2E fixture for the MCP tool approval gate — a stub MCP server with one read and one gated mutation, gated by the server's archie policy in .mcp.json. Used by tools/e2e/tool-gate-check.ts; harmless to leave installed.

sweatco/archie-hq

Plugin Claude Code

Example domain plugin — a generic assistant agent that summarizes text and drafts short responses. A minimal, runnable starting point for building your own plugins.

sweatco/archie-hq

Plugin Claude Code

PM extension — example orchestration skill and a minimal PM overlay. Shows how the PM agent is extended with business context and workflows.

Raven-Scout/scout-plugin

Plugin Claude Code

Marketplace for the Scout autonomous knowledge management plugin.

Raven-Scout/scout-plugin

Plugin Claude Code

Autonomous knowledge management and daily briefing system. Monitors your work tools (Slack, Calendar, Linear, GitHub, etc.), synthesizes findings into a persistent knowledge base with a formal ontology, and delivers daily action items — all running unattended via scheduled Claude Code sessions. Six session types…

a-ariff/ariff-claude-plugins

Plugin Claude Code

Plugin marketplace listing 65 plugins: analyzer, answer-analyzer, answer-validator, anti-hallucination, api-docs-generator.

a-ariff/ariff-claude-plugins

Plugin Claude Code

Code analysis and metrics specialist. Use for codebase analysis, complexity metrics, dependency analysis, and generating insights about code quality and structure.

a-ariff/ariff-claude-plugins

Plugin Claude Code

System architecture and design specialist. Use for designing scalable systems, making architectural decisions, creating technical specifications, or refactoring system structure.

bonfire-systems/goalkeeper

Plugin Claude Code

Durable goal-following for Claude Code: contracts with definition-of-done, subagent judge gates, and linear chains.

bonfire-systems/goalkeeper

Plugin Claude Code

Durable goal-following for Claude Code: contracts with definition-of-done, subagent judge gates, executor-subagent chain execution, and adaptive missions — with all state mechanics executed by a tested CLI and invariants enforced by hooks. Inspired by OpenAI Codex /goal and the Ralph loop pattern.

krokoko/cairn

Plugin Claude Code

Plugin marketplace listing 2 plugins: codebase-ai-readiness, software-verification.

krokoko/cairn

Plugin Claude Code

Assess how AI-friendly a codebase is and produce an autonomy maturity map with scores, blockers, and a roadmap.

krokoko/cairn

Plugin Claude Code

Design verification strategies that unlock higher levels of autonomous software development.

proyecto26/autoresearch-ai-plugin

Plugin Claude Code

Autonomous experiment loops for Claude Code — edit, measure, keep or discard, repeat forever.

proyecto26/autoresearch-ai-plugin

Plugin Claude Code

Autonomous experiment loop for Claude Code. Run an endless optimize-measure-keep/discard cycle on any optimization target: LLM training loss, test speed, bundle size, build time, and more.

watt-mind/factory

Plugin Claude Code

Plugin marketplace listing 1 plugin: core.

watt-mind/factory

Plugin Claude Code

Watt Mind execution protocol: dispatch, review/merge, triage, and conventions audit.

dork-labs/dorkos

Plugin Claude Code

The /flow engine — one unified, PM-agnostic workflow system spanning capture → done. Manual stages and an autonomous loop seated on DorkOS Pulse.

open-covenant/covenant

Plugin Claude Code

Covenant plugins for Claude Code.

open-covenant/covenant

Plugin Claude Code

Run Claude Code unattended: cap the spend, sandbox execution, and get a signed receipt, enforced from outside the agent's own process by the covguard binary.