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/vassiliylakhonin/agenda-intelligence-mdWrote 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/vassiliylakhonin/agenda-intelligence-md/output-verification)<a href="https://agentmods.dev/plugins/vassiliylakhonin/agenda-intelligence-md/output-verification"><img src="https://agentmods.dev/badge/plugins/vassiliylakhonin/agenda-intelligence-md/output-verification/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/vassiliylakhonin/agenda-intelligence-md/output-verification"><img src="https://agentmods.dev/badge/plugins/vassiliylakhonin/agenda-intelligence-md/output-verification.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
output-verification 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 today.
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": "output-verification",
"version": "1.16.2",
"description": "Review structural gaps in caller-supplied claim evidence. No independent source verification; every verdict requires human review.",
"author": {
"name": "Vassiliy Lakhonin",
"url": "https://github.com/vassiliylakhonin"
},
"homepage": "https://agent-output-verification-a2a.vassiliy-lakhonin.workers.dev/",
"repository": "https://github.com/vassiliylakhonin/agenda-intelligence-md",
"license": "MIT",
"mcpServers": {
"output-verification": {
"type": "http",
"url": "https://agent-output-verification-a2a.vassiliy-lakhonin.workers.dev/mcp/output-verification"
}
}
}
What it installs
The manifest is a name and a version. 1 skill travel with it, and installing the plugin installs all of them — 25 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.
- today Changed dbae78451671
- yesterday Changed f9485e5fb1d8
- 6d ago Changed e8cc0f5171a7
- 7d ago Changed bbac48015b93
- 15d ago Changed f129c3fc015a
- 19d ago First seen · 19 lines scan A 9e4108979ba0
output-verification is a plugin published in the GitHub repository vassiliylakhonin/agenda-intelligence-md (8 stars, last pushed yesterday), 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 2026-09-28.
Other plugins, from other repositories
deepeval
Skills for adding DeepEval evaluations, tracing, datasets, Confident AI reports, and iterative improvement loops to AI applications.
artel
Ambient shared memory for Claude Code: injects the note you need when you need it — your last handoff at session start, relevant notes on each prompt, file-anchored gotchas before you edit — and writes the session back afterwards, including what it cost. Points at your own self-hosted Artel via ARTELURL / ARTELAGENT.
mnemo-mcp
Persistent AI memory — store, search, and recall knowledge across sessions.
althing
Run synthetic focus groups and user research panels using AI personas. Define personas in YAML, design survey instruments, and collect structured qualitative feedback — all from Claude Code.
pyeye
Semantic Python code analysis MCP server - find symbols, navigate definitions, trace call hierarchies, and understand code structure.
tempguru-event-staffing
W-2 event staffing data for AI agents: 300+ U.S. and Canadian markets, roles, rates, lead times, compliance. Twelve MCP tools: the call-first planstaffing planner, read-only lookups, two plan-persistence tools, and the requestquote form handoff, no auth.