agentrigs
+ submit rig
~ / rigs / jajupmochi / agent-harness

jajupmochi/agent-harness

Linlin's curated AI agent harness configuration: workflow rules, skills, hooks, plugin recommendations, tooling preferences, and project templates. Loadable into any new project so a fresh /init can pick the relevant subset.

↗ GitHub ★ 1 mit updated 3mo ago personal setup Claude CodeCodex Claude plugin
share on X
ARCHETYPE
Skill Collector
Ten-plus skills loaded on demand. A procedural-knowledge library.
CONTEXT TAX · EVERY TURN
~3.3k tokens
Moderate · median rig: 2.3k · breakdown
GUARDRAILS
0/5
No committed guardrails · details

Copy this rig

# review before running: this installs third-party code
$ npx degit jajupmochi/agent-harness/.claude ./rig-agent-harness  # inspect, then merge into .claude/

MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.

$ npx degit jajupmochi/agent-harness/.claude/skills/new-hook .claude/skills/new-hook
$ npx degit jajupmochi/agent-harness/.claude/skills/new-rule .claude/skills/new-rule
$ npx degit jajupmochi/agent-harness/.claude/skills/new-skill .claude/skills/new-skill
$ npx degit jajupmochi/agent-harness/.claude/skills/publish .claude/skills/publish
$ npx degit jajupmochi/agent-harness/skills/agent-config-adapter .claude/skills/agent-config-adapter
$ npx degit jajupmochi/agent-harness/skills/agent-update-watcher .claude/skills/agent-update-watcher
$ npx degit jajupmochi/agent-harness/skills/autopilot .claude/skills/autopilot
$ npx degit jajupmochi/agent-harness/skills/autoresearch-toolfinder .claude/skills/autoresearch-toolfinder
$ npx degit jajupmochi/agent-harness/skills/claude-desktop-sessions .claude/skills/claude-desktop-sessions
$ npx degit jajupmochi/agent-harness/skills/code-verifier .claude/skills/code-verifier
$ npx degit jajupmochi/agent-harness/skills/doc-writing .claude/skills/doc-writing
$ npx degit jajupmochi/agent-harness/skills/end-of-turn-marker .claude/skills/end-of-turn-marker
$ npx degit jajupmochi/agent-harness/skills/figma-authoring-constraints .claude/skills/figma-authoring-constraints
$ npx degit jajupmochi/agent-harness/skills/figma-design-fetch .claude/skills/figma-design-fetch
$ npx degit jajupmochi/agent-harness/skills/general .claude/skills/general
$ npx degit jajupmochi/agent-harness/skills/init-codex-config .claude/skills/init-codex-config
$ npx degit jajupmochi/agent-harness/skills/long-running-tasks .claude/skills/long-running-tasks
$ npx degit jajupmochi/agent-harness/skills/memory-flywheel .claude/skills/memory-flywheel
$ npx degit jajupmochi/agent-harness/skills/preview-template .claude/skills/preview-template
$ npx degit jajupmochi/agent-harness/skills/privacy-redact .claude/skills/privacy-redact
$ npx degit jajupmochi/agent-harness/skills/prompt-library .claude/skills/prompt-library
$ npx degit jajupmochi/agent-harness/skills/research-critic .claude/skills/research-critic
$ npx degit jajupmochi/agent-harness/skills/system-cleanup .claude/skills/system-cleanup
$ npx degit jajupmochi/agent-harness/skills/task-orchestrator .claude/skills/task-orchestrator
$ npx degit jajupmochi/agent-harness/skills/ta<redacted> .claude/skills/ta<redacted>
$ npx degit jajupmochi/agent-harness/skills/tui-installer .claude/skills/tui-installer
$ npx degit jajupmochi/agent-harness/skills/verify-template .claude/skills/verify-template
$ npx degit jajupmochi/agent-harness/skills/verify-visual .claude/skills/verify-visual
$ npx degit jajupmochi/agent-harness/templates/personal-cite-static/.claude/skills/i18n-sync .claude/skills/i18n-sync
$ npx degit jajupmochi/agent-harness/templates/personal-cite-static/.claude/skills/preview .claude/skills/preview
$ npx degit jajupmochi/agent-harness/templates/research-package-py/.claude/skills/verify .claude/skills/verify

This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 6,974 rigs:

{
  "permissions": {
    "deny": [
      "Read(./.env)",
      "Read(**/.env)",
      "Read(~/.ssh/**)",
      "Bash(rm -rf *)",
      "Read(**/*.pem)",
      "Bash(rm -rf /)",
      "Bash(git push --force:*)",
      "Bash(sudo *)",
      "Read(~/.aws/**)",
      "Bash(rm -rf /*)",
      "Read(./.env.*)",
      "Read(.env)",
      "Bash(git push --force*)",
      "Bash(rm -rf:*)",
      "Read(**/.env.*)",
      "Read(**/*.key)",
      "Bash(sudo:*)",
      "Bash(git reset --hard*)",
      "Bash(git reset --hard:*)",
      "Read(.env.*)"
    ],
    "ask": [
      "Bash(git push:*)",
      "Bash(git push *)",
      "Bash(git commit:*)",
      "Bash(rm *)",
      "Bash(rm:*)",
      "Bash(npm publish:*)",
      "Bash(wget *)",
      "Bash(git rebase *)",
      "Bash(gh pr merge *)",
      "Bash(git commit *)"
    ]
  }
}

Skills (31)

new-hooknew-rulenew-skillpublishagent-config-adapteragent-update-watcherautopilotautoresearch-toolfinderclaude-desktop-sessionscode-verifierdoc-writingend-of-turn-markerfigma-authoring-constraintsfigma-design-fetchagent-harness-generalinit-codex-configlong-running-tasksmemory-flywheelpreview-templateprivacy-redactprompt-libraryresearch-criticsystem-cleanuptask-orchestratorta<redacted>tui-installerverify-templateverify-visuali18n-syncpreviewverify

Similar rigs

jongwony/epistemic-protocols
Epistemic protocols for Claude Code — structure human-AI interaction quality at every decision point - https://epistemic-protocols.com
★ 170 project Claude CodeCodex 29 skills 1 hooks 9d ago
Skill Collector 6.8k tok ·
zenbu-labs/terminal-browser
A browser inside your terminal
★ 3,734 project Claude CodeCodex 1 skills 2d ago
Minimalist 815 tok ·
cosmtrek/mindwalk
A visualization tool that replays coding-agent sessions on a 3D map of your codebase.
★ 1,364 project Claude CodeCodex 1 skills 2mo ago
Minimalist 1.1k tok ·
alleriasun/clearclaw
—
★ 625 project Claude Code 1 skills today
Minimalist 1.0k tok ·
copied ✓