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.
git clone --depth 1 https://github.com/serdardb/multi-agentWrote 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.
[](https://agentmods.dev/plugins/serdardb/multi-agent/plugin)<a href="https://agentmods.dev/plugins/serdardb/multi-agent/plugin"><img src="https://agentmods.dev/badge/plugins/serdardb/multi-agent/plugin/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.
<a href="https://agentmods.dev/plugins/serdardb/multi-agent/plugin"><img src="https://agentmods.dev/badge/plugins/serdardb/multi-agent/plugin.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
multi-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 1mo 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.
What it actually says
{
"name": "multi-agent",
"version": "0.1.2",
"description": "Run your OWN agent on other AI model CLIs (grok / codex / claude) as independent second opinions — no API, each model runs the agent natively. Two commands: /multi-agent (a single-model, live, steerable review you can talk to while it runs) and /arena (a deterministic multi-model adversarial arena — independent audit → cross-critique until positions converge → code-verifying judge → scoreboard, shown live in /workflows).",
"author": { "name": "Serdar Doğrubakar", "url": "https://github.com/serdardb" },
"homepage": "https://github.com/serdardb/multi-agent",
"repository": "https://github.com/serdardb/multi-agent",
"license": "MIT",
"keywords": ["code-review", "multi-model", "cross-model", "arena", "adversarial", "second-opinion", "security", "claude-code", "grok", "codex", "workflow"]
}
What it installs
The manifest is a name and a version. 2 skills, 3 agents travel with it, and installing the plugin installs all of them — 197 tokens a session between them. Each is measured on its own page, and each can be installed alone.
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.
- 1mo ago First seen · 11 lines scan A dd40abcd973a
multi-agent is a plugin published in the GitHub repository serdardb/multi-agent (1 stars, last pushed 2mo 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 2026-08-31.
Other plugins, from other repositories
code-gauntlet
Adversarial code review for Claude Code: each GitHub PR or GitLab MR runs a gauntlet of up to seven specialist agents for bugs, security, tests and cross-file impact, and every finding must survive verification, a skeptical validator, and a blind challenge before it is posted.
codebase-health
Codebase health tools: deep audits via orchestrated specialists (security, performance, libraries, quality, dead code), safe dependency updates, dead-code cleanup, and documentation maintenance.
research-assistant
Research libraries, frameworks, and APIs using official documentation. Includes /research for comprehensive investigation and /official-docs for pre-task doc lookups. Works best with Context7 and Fetch MCP servers installed (falls back to WebSearch/WebFetch if unavailable).
rn-agents-kit
Audit-first agent skills for React Native maintenance: code-quality audit, New Architecture readiness, asset hygiene, project snapshot, device screen capture, Metro console capture, runtime QA routing, CI artifact inventory, EAS build profile resolution, and upgrade surface assessment.
taxcraft
US tax and accounting work for people whose situation outgrew consumer software but who still do the prep themselves: multiple entities, K-1s that arrive late and wrong, basis and carryforwards nobody is tracking, quarterly estimates guessed at rather than computed. It builds the workpapers a CPA would build …
craftsman
Ship something with Lovable, Replit, Bolt, v0, Claude Code, or Codex and it works, but nobody with an engineering background has checked it. This is the engineering review you'd get from the technical co-founder you don't have. craft-audit checks the whole project across ten areas: UX, frontend, backend, database…