claude

autoresearch

Autonomous experiment loop for ZCode: set a goal, pick a mechanical metric, and let the agent iterate. It measures, keeps what works, reverts what doesn't, and repeats. Provides init/run/log experiment tools (MCP), a loop protocol skill, guardrails (frozen benchmark, checks backpressure, memory injection, Stop continuation), and static + live dashboards.

0 on GitHubby luochang212v0.1.0Apache-2.0updated 10 days ago

Not in a marketplace yet, so it loads for one session with --plugin-dir.

Plugin autoresearch in kael-odin/zcode-plugins-zh. Commands read from the repo's manifest — 9 days ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use and it starts local MCP servers. The exact commands are listed below; read them before installing.

What it installs

Skills (2)

  • autoresearch-hooks

    Teach how to write iteration hooks (before/after scripts) for the autoresearch plugin: contract, scenario selection, mock testing, and guidelines. Use when the user wants to add a custom hook, extend the loop with external logic (lookups, notifications, journals, guards), or understand the .auto/hoo

  • autoresearch

    Run an autonomous experiment loop in ZCode. Set a goal and a mechanical metric, then iterate: modify code, run the benchmark, keep improvements, revert regressions, repeat. Use when the user wants the agent to autonomously optimize something measurable (runtime, bundle size, test speed, loss, any nu

Slash commands (5)

  • /autoresearch:autoresearch

    Enter autoresearch mode. Resumes from .auto/prompt.md when a session exists, otherwise sets one up from your goal. Usage: /autoresearch:autoresearch <goal or resume context>

  • /autoresearch:clear

    Clear the autoresearch session by deleting .auto/log.jsonl, then start fresh. Keeps measure.sh / checks.sh / prompt.md. Usage: /autoresearch:clear

  • /autoresearch:export

    Render .auto/log.jsonl into autoresearch-dashboard.html. Usage: /autoresearch:export

  • /autoresearch:finalize

    Split kept experiments into clean topic branches you can PR. Usage: /autoresearch:finalize

  • /autoresearch:off

    Stop the auto-resume hints while keeping the session. Resume anytime with /autoresearch:autoresearch. Usage: /autoresearch:off

MCP servers (1)

  • autoresearch

    runs: node ${ZCODE_PLUGIN_ROOT}/mcp/server.ts

Hooks (5)

  • Stop

    node

  • PreToolUse

    node

  • PermissionRequest

    node

  • UserPromptSubmit

    node

  • SessionStart

    node

View source on GitHub kael-odin.github.io

More from kael-odin/zcode-plugins-zh

All 54 plugins →

cloudbase-skills

claudecodex

腾讯云 CloudBase 开发技能与 MCP 集成，覆盖 Web、微信小程序、数据库、云函数、云托管、云存储和 AI 项目的开发、部署与排障。

30 skills · 1 MCP server

0

write-research

claude

End-to-end investment research reports, sector analysis, earnings updates, and valuation models

17 skills · 13 commands · 1 agent

0

run-fpa

claude

Corporate finance and FP&A: 管理口径 multi-dimensional management reporting off a closed ledger, rolling cash-flow forecasts, budget-versus-actual variance analysis, scenario and break-even analysis off a named base case, incremental business-case support for investment and cost decisions, and peer benchmarking against listed comparables

10 skills · 8 commands · 1 agent

0

accounting-and-reporting

claudecodex

【企业财务】核算与报告:月结关账检查与阻断项、总账与明细账勾稽及交易级差异归因、科目映射与重分类、内部管理报表编制、三表勾稽复核

8 skills · 5 commands · 1 agent

0

pick-funds

claude

Fund and fund-manager research: multi-criteria fund screening, fund and manager profiles, holdings and style analysis, and shortlist comparisons for funds, ETFs, and LOFs

7 skills · 5 commands · 1 agent

0

video2code

claudecodex

基于 ZCode 内置 Browser Use WebView 录制 WebM 并用 ffmpeg 转成 MP4/URL 复刻；无需 Playwright 或外部 Chromium

6 skills · 4 commands · 2 MCP servers · hooks

0

Similar plugins

ralph-loop

claude

Interactive self-referential AI loops for iterative development. Claude works on the same task repeatedly, seeing its previous work, until completion.

3 commands · hooks

anthropics/claude-code 155,476

nextjs

claude

Official Next.js skills: adopt and optimize Cache Components, adopt Partial Prefetching, and verify runtime behavior against a running dev server.

6 skills

vercel/next.js 143,495

ruflo-cost-tracker

claude

Token usage tracking, model cost attribution per agent, budget alerts, and optimization recommendations

24 skills · 1 command · 1 agent · hooks

ruvnet/ruflo 81,589

trackio-cli

claude

Query Trackio projects, runs, and metrics using the `trackio` CLI. Use when the user needs to list projects/runs/metrics, get summaries, or retrieve metric values from local Trackio databases. Supports both human-readable and JSON output formats. Covers project discovery, run inspection, metric querying, and system metrics. Designed for LLM agents and automation scripts.

1 skill

sickn33/agentic-awesome-skills 49,620

cloud-infrastructure

claudecodex

Cloud architecture design for AWS/Azure/GCP/OCI, Kubernetes cluster configuration, Terraform infrastructure-as-code, hybrid cloud networking, and multi-cloud cost optimization

8 skills · 7 agents

wshobson/agents 42,461

Zoom

claude

Plan, build, and debug Zoom integrations across REST APIs, Meeting SDK, Video SDK, webhooks, bots, and MCP workflows. Search meetings, retrieve recordings, access transcripts, and design AI-powered Zoom experiences.

44 skills · 3 MCP servers

anthropics/knowledge-work-plugins 35,601

Maintain autoresearch?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins54 plugins