agentmods.dev

agent

agent is a plugin for Claude Code from air-gapped/skills. Its manifest loads nothing; the 2 skills it bundles cost 189 tokens per session together, scanned A, original, MIT.

A toolkit for improving coding-agent workflows through repeated experiments and skill refinement. Autoresearch uses measured trial-and-error changes, while a skill-improvement loop evaluates and adjusts agent instructions.

In plain words
What is it for?
Use it to run hill-climbing experiments, coordinate multi-agent research, evaluate workflow changes, and refine Claude Code skills.
Why use it?
It helps when an agent workflow needs to improve based on evidence rather than one-off manual edits. The included research agents can support investigation and scoring of alternatives.

Plugin for Claude Code

Written for Claude Code: a Claude Code plugin manifest. Also seen: mentions Claude Code.

not rated 5repo changed 12d ago A scan tokens not measured original MIT

Good fit Use it to run hill-climbing experiments, coordinate multi-agent research, evaluate workflow changes, and refine Claude Code skills.

Compare 6 plugins from other repositories ↓
Install in Claude Code
/plugin marketplace add air-gapped/skills
/plugin install agent
View source ↗ air-gapped/skills
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/air-gapped/skills

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for agent

README.md
[![agentmods](https://agentmods.dev/badge/plugins/air-gapped/skills/agent/github.svg)](https://agentmods.dev/plugins/air-gapped/skills/agent)
Your own site
<a href="https://agentmods.dev/plugins/air-gapped/skills/agent"><img src="https://agentmods.dev/badge/plugins/air-gapped/skills/agent/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for agent

Your own site · 80×15
<a href="https://agentmods.dev/plugins/air-gapped/skills/agent"><img src="https://agentmods.dev/badge/plugins/air-gapped/skills/agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session not measured What this adds to a session before it is invoked.
When invoked not measured The manifest loads nothing itself; its 2 skills cost 189 tokens a session between them.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Security

Grade A, and why

agent scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude-plugin/marketplace.json#agent · 29 lines

What it actually says

{
  "name": "agent",
  "source": "./",
  "description": "Agent workflow suite — Karpathy-pattern autoresearch (hill-climbing, multi-agent research) and skill-improver (Claude Code skill optimization loop).",
  "version": "0.20261004.216",
  "author": {
    "name": "Jörgen"
  },
  "license": "MIT",
  "skills": [
    "./.claude/skills/autoresearch",
    "./.claude/skills/skill-improver"
  ],
  "strict": false,
  "agents": [
    "./.claude/agents/blind-scorer.md",
    "./.claude/agents/deep-researcher.md",
    "./.claude/agents/web-searcher.md"
  ],
  "category": "ai-workflow",
  "tags": [
    "autoresearch",
    "agents",
    "claude-code",
    "skills",
    "hill-climbing",
    "optimization"
  ]
}
Contents

What it installs

The manifest is a name and a version. 2 skills travel with it, and installing the plugin installs all of them — 189 tokens a session between them. Each is measured on its own page, and each can be installed alone.

Files

What ships with it

1 file beside marketplace.json#agent in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago Changed a16d11735be0
  2. 13d ago Changed add1e8401a10
  3. 23d ago Changed 5b13fc626767
  4. 24d ago Changed 52f92a2b08d7
  5. 1mo ago Changed 839cb2752468
  6. 1mo ago Changed 934a07671249
  7. 1mo ago First seen · 29 lines scan A 8f3442b324b9

Subscribe to this mod's changes

agent is a plugin published in the GitHub repository air-gapped/skills (5 stars, last pushed 2d ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Building agentsClaude CodeResearchagent-skillsai-agentsair-gappedanthropicclaude-code-marketplace
Related

Other plugins, from other repositories

issue-to-pr

Take a GitHub issue (bare, or tracked on a Project board) from triage to a merge-ready PR through a gated, worktree-isolated pipeline that scales its depth to the task and asks at most one batched question. The gates hold every run: design hardening, tests green, code review clean. Once you approve the PR in-session…

DmitriyYukhanov/claude-plugins · not measured

claude-md-slim

Shrink an oversized or stale CLAUDE.md by moving subsystem detail into path-scoped .claude/rules/ files, with scripts that measure the loaded size against the documented 200-line budget and prove the extraction lost nothing.

DmitriyYukhanov/claude-plugins · not measured

humanizer

Remove AI-generation patterns from English or Russian text — auto-detects language and applies a matching MIT-licensed ruleset (33-pattern EN from blader/humanizer, 54-pattern RU from ilyautov/humanizer-ru) with explicit override and mixed-text handling.

DmitriyYukhanov/claude-plugins · not measured

agent-dispatch

Start issue-to-pr runs from GitHub labels on your own machine: a scheduled tick picks the issue you labelled agent, or the one you replied on, and runs it headless under your own Claude Code or Codex login, on Windows, macOS or Linux.

DmitriyYukhanov/claude-plugins · not measured

agent-teams

Set up and orchestrate Claude Code agent teams - multiple instances working together with shared tasks, messaging, and coordination.

DmitriyYukhanov/claude-plugins · not measured

codex-collaboration

Cross-model collaboration between Claude and Codex — sequential drive/validate/act loops and parallel dual review with triage.

DmitriyYukhanov/claude-plugins · not measured