70 plugins from public Claude Code marketplaces. Each one bundles skills, slash commands, agents, hooks or MCP servers that install together.
Showing 70 plugins that match these filters.
research-agent-skillsKalarisLabs/research-agent-skills2
claude-code-workflowswshobson/agents1
pskoett-skillspskoett/pskoett-ai-skills1
autoharnesstigerless-labs/autoharness1Submit the repo with .claude-plugin/marketplace.json and we list every plugin.
Submit marketplaceEval-gated LLM fine-tuning lifecycle (LoRA/QLoRA, DPO, GRPO/RLVR, vision SFT, quantized export). Use when fine-tuning an open-weights model end-to-end: building the eval harness first, selecting method and model, preparing data, training with Unsloth or TRL, gating checkpoints, and exporting GGUF/FP8.
Build, launch, grade, and schedule Claude Managed Agents (CMA) in your own Anthropic account — a plugin re-implementation of Anthropic's launch-your-agent reference skill (Apache-2.0; independent, not a fork). Every session starts with a goal (./my-agent/goal.json, opt-in SessionStart hook, driven by /cs-goal); loop_compiler.py compiles it into a bounded grade->iterate loop (max_iterations 1..20), a recurring cron scheduled-deployment loop, or a single-pass interview->stage->launch workflow. 6 skills: orchestrator (context: fork goal router) + interview + stage-launch + grade-iterate + run-without-you + wrap-up. 18 stdlib-only deterministic scaffolders (no network/API calls; live launches emitted as BYOK curl that never prints the key), 4 agents, 8 commands, hooks, 5 references, 4 assets. Validators enforce CMA limits. Distinct from engineering/agent-harness (generic domain loop) and write-a-skill (authors Claude Code skills, not CMAs).
Two-loop skill pipeline for Claude Code. Inner loop (plan-interview, intent-framed-agent, session-context, context-surfing, verify-gate, self-healing, simplify-and-harden) catches failures in-session — including active runtime recovery via self-healing when a command, test, build, helper, env, or external service fails mid-task. Outer loop (learning-aggregator, self-improvement, eval-creator, pre-flight-check) captures recurring patterns, promotes them to CLAUDE.md as permanent rules with human approval, and turns failures into regression tests. The harness gets stronger with every cycle.
Turn any domain folder of skills into a bounded agentic loop: a manifest builder inventories a domain's skills/tools/checks, a goal compiler turns a goal into a verifiable task plan (refusing vague goals with forcing questions), and a JSON-backed loop controller drives execute->verify->close with retry caps, controller-run verification (no verification theater), human escalation on exhausted budgets, and a close gate that refuses while any task is unverified or unwaived. Ships 3 stdlib Python tools, 18 committed per-domain harness manifests + JSON schema, 3 references citing the 2024-2026 agent-harness canon (Anthropic long-running harnesses, verifier's law, SWE-agent, Ralph loop, Cognition), harness-runner agent + /cs-harness command. Use when an agent or subagent should pick up a goal, define its tasks, complete and verify them, and close the loop.
Per-symbol maintenance optimizer for an agent's skill layer — learns skills from real runs, keeps them by adherence in use.
Evaluation harnesses, interpretability and safety/alignment.
researchAutonomous research loops, agent harnesses and research artifacts.
researchLoops (a file-based knowledge base) + a codebase harness (dev-local, e2e, crabbox cloud boxes, verify-before-ship) so agents can run, test, and ship code autonomously.
engineeringEnterprise architecture skill pack (requirements -> design -> validate -> enforce gate -> review/optimize/report)
MOOS-IvP skills for apps, IvP behaviors, missions, test harnesses, documentation, and alog analysis.
developer toolsYour coding agent already plans and swarms — Wicked Garden is the curated toolkit for what it can't: proof-not-claims gates, code links grep can't see, deterministic refactor, cross-session memory, multi-model second opinions (fallback seats labelled). Don't fight the harness; fill its gaps. The gate needs wicked-vault (a self-contained infra peer, installed directly) and re-derives through garden's built-in engine; estate/bus are opt-in layers (wicked-testing and wicked-brain are retired — QE ships in-catalog as the qe domain, memory lives in wicked-estate).
AI with a harness. Agent skills, rules, CI and security gates for AI-assisted development — a diagnose → issue → plan → resolve ⇄ verify loop, shipped as one harness.
Security find-and-fix loop — adapted (Apache-2.0) from anthropics/defending-code-reference-harness and generalised to all vulnerability classes. Four read/write-only skills (no target-code execution): threat-model (bootstrap from code+CVEs+git-history, or interview an owner via Shostack's four questions, or both -> THREAT_MODEL.md), vuln-scan (static source review, parallel per-focus-area subagents -> VULN-FINDINGS.json), triage (adversarial N-vote verification, 16 false-positive exclusion rules, root-cause dedup, impact-on-asset x exploitability severity with stated deployment conditions, owner routing, checkpoint/resume -> TRIAGE.json), and patch (root-cause candidate diffs + independent reviewer, fix-first priority for deployment-conditional dangerous primitives, never auto-applied -> PATCHES/). The autonomous C/C++ + AddressSanitizer execution pipeline this was extracted from is referenced, not bundled (vuln-scan/HARNESS.md).
securityGoogle Agent Development Kit (ADK) starter for building production AI agents — ReAct single-agent or multi-agent orchestration (Sequential/Parallel/Loop), tool wiring, evaluation harness, and optional Vertex AI Agent Engine deployment.
Web E2E testing with Playwright, Cypress, Selenium and WebdriverIO, plus cloud grids, a self-hosted Selenium Grid 4, and cross-browser test matrices. Covers BrowserStack, Sauce Labs and LambdaTest, traffic-based T1/T2/T3 browser tiers, a framework decision guide, and Playwright's agent tooling (playwright-cli, Playwright MCP, planner/generator/healer Test Agents). Agents (playwright-codegen-reviewer, automation-harness-bootstrapper) refactor recorded Playwright or Cypress Studio specs and scaffold a full E2E harness for a repo with no suite.
functional-testingMobile app testing: XCUITest and Espresso for native apps, Appium, Detox and Maestro across platforms, Flutter tests, and mobile accessibility checks. XCUITest also covers macOS desktop apps, and a tiered device matrix keeps CI cost down. An authoring agent (mobile-test-author) detects the driver, scaffolds a harness when none exists, and writes one test per spec. Mobile-web viewport emulation lives in qa-web-e2e.
functional-testingFeature-flag and experimentation testing: LaunchDarkly and OpenFeature SDK tests, flag test matrices, A/B test validity, and experiment result analysis. Covers Unleash, Flagsmith and GrowthBook through OpenFeature, and hermetic harnesses for Statsig, Optimizely, Split.io, Amplitude Experiment and VWO. Agents (flag-coverage-gap-detector, stale-flag-detector, sample-ratio-mismatch-detector) find untested flag branches, stale flags, and sample ratio mismatch. Distinct from qa-test-environment/feature-flag-test-harness and qa-shift-right/feature-flag-experiment-validator.
integrations-protocolsWarn-only, zero-dependency runtime harness layer for Claude Code (companion to superpowers): persistent state, scope control (prompt + action side), Default-FAIL verification & coverage gates, entropy governance (golden-rules + drift detection), and first-class C/C++ engineering (CMake/Meson/Make/Bazel + clang-tidy + sanitizers).
Embedded C/C++ testing: GoogleTest on ARM, Unity and Ceedling with CMock mocks, QEMU system emulation, gcov coverage, and hardware-in-the-loop. An authoring agent (embedded-test-author) detects the framework and execution path, scaffolds the harness when none exists, and writes one unit test per spec.
functional-testingSelf-contained agentic development harness: one command runs the full feature pipeline with double review and security gates.
