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.
/plugin marketplace add moberghr/mtk-agent-toolkitnpx agentmods add plugins/moberghr/mtk-agent-toolkit/mtk-agent-toolkitgit clone --depth 1 https://github.com/moberghr/mtk-agent-toolkitWrote 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/moberghr/mtk-agent-toolkit/mtk-agent-toolkit)<a href="https://agentmods.dev/plugins/moberghr/mtk-agent-toolkit/mtk-agent-toolkit"><img src="https://agentmods.dev/badge/plugins/moberghr/mtk-agent-toolkit/mtk-agent-toolkit.svg" alt="Measured on agentmods" height="20"></a>Grade B, and why
mtk scanned grade B with 1 finding 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 3d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
"settings.json": "./.claude/settings.json" What it actually says
{
"name": "mtk",
"description": "MTK (Moberg Toolkit) — AI-assisted development toolkit with pluggable tech stacks (.NET, Python, TypeScript). Language-agnostic workflow skills, review agents, and compliance-first workflows.",
"version": "7.33.0",
"author": {
"name": "Moberg d.o.o.",
"url": "https://www.moberg.hr"
},
"homepage": "https://github.com/moberghr/mtk-agent-toolkit",
"repository": "https://github.com/moberghr/mtk-agent-toolkit",
"license": "MIT",
"keywords": [
"dotnet",
"python",
"typescript",
"react",
"nextjs",
"tauri",
"serious-software",
"regulated-software",
"compliance",
"code-review",
"implementation-workflow",
"skills",
"tech-stack"
],
"skills": "./.claude/skills",
"agents": [
"./.claude/agents/architecture-reviewer.md",
"./.claude/agents/compliance-reviewer.md",
"./.claude/agents/test-reviewer.md",
"./.claude/agents/silent-failure-hunter.md",
"./.claude/agents/plan-gap-reviewer.md",
"./.claude/agents/context-miner.md"
],
"settings.json": "./.claude/settings.json"
}
What it installs
The manifest is a name and a version. 45 skills, 6 agents, 7 hooks, 1 MCP server travel with it, and installing the plugin installs all of them — 1,762 tokens a session between them. Each is measured on its own page, and each can be installed alone.
- Skill setup-audit A 27 tokens
- Skill spec-driven-development A 31 tokens
- Skill tech-stack-typescript A 43 tokens
- Skill mtk A 17 tokens
- Skill tech-stack-dotnet A 29 tokens
- Skill tech-stack-python A 27 tokens
- Skill code-review-and-quality A 35 tokens
- Skill mtk-setup A 0 tokens
- Skill promote-lesson A 42 tokens
- Skill research-context A 40 tokens
- Skill spec-drift-detection A 39 tokens
- Skill claude-md-audit A 45 tokens
- Skill claude-md-capture A 43 tokens
- Skill planning-and-task-breakdown A 29 tokens
- Skill pre-commit-review A 21 tokens
- Skill prior-work-check A 33 tokens
- Skill handoff A 42 tokens
- Skill repo-health A 28 tokens
- Skill workflow-artifacts A 36 tokens
- Skill incremental-implementation A 29 tokens
- Skill mtk-doctor A 0 tokens
- Skill using-git-worktrees A 29 tokens
- Skill security-and-hardening A 37 tokens
- Skill debugging-and-error-recovery A 36 tokens
- Skill pr-review-mining A 31 tokens
- Skill subagent-implementation A 46 tokens
- Skill lesson-mining A 30 tokens
- Skill brainstorming A 29 tokens
- Skill writing-skills A 35 tokens
- Skill code-simplification A 30 tokens
- Skill lesson-refresh A 44 tokens
- Skill test-driven-development A 28 tokens
- Skill source-driven-development A 39 tokens
- Skill batch-fix B 41 tokens
- Skill implement B 17 tokens
- Skill verification-before-completion B 27 tokens
- Skill context-engineering B 32 tokens
- Skill setup-refresh B 42 tokens
- Skill fix B 42 tokens
- Skill golden-path-capture B 43 tokens
- Skill correction-capture B 39 tokens
- Skill setup-converge B 37 tokens
- Skill toolkit-health B 40 tokens
- Skill context-report B 32 tokens
- Skill setup-bootstrap D 30 tokens
- Agent compliance-reviewer A 48 tokens
- Agent plan-gap-reviewer A 35 tokens
- Agent silent-failure-hunter A 44 tokens
- Agent architecture-reviewer A 20 tokens
- Agent test-reviewer A 17 tokens
- Agent context-miner A 57 tokens
- Hook PreToolUse A not measured
- Hook Stop A not measured
- Hook SessionStart A not measured
- Hook SubagentStop A not measured
- Hook UserPromptSubmit A not measured
- Hook PostToolUse A not measured
- Hook PreCompact A not measured
- MCP server mtk-context A not measured
What ships with it
1 file beside plugin.json 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.
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.
- 3d ago First seen · 38 lines scan B 11e027bd1dba
mtk is a plugin published in the GitHub repository moberghr/mtk-agent-toolkit (7 stars, last pushed 9d 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 B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other plugins, from other repositories
specter
EU AI Act compliance toolkit for Claude Code — validate citations against the canonical 113-article + 13-annex catalog, query the four-axis agentic compound-risk taxonomy, look up role obligations, and run the LLM-as-Judge reward-hack detector on proposed remediation tasks. Wraps the specter Python package as a local.
agb-recht-pruefer
Gigantischer AGB-Rechtsprüfer und Klausel-Entwerfer für deutsches Recht: §§ 305 bis 310 BGB, UKlaG, B2C/B2B, Branchen-AGB, Redlining, Klauselrisiko und rechtssichere Entwurfsworkflows.
go-ultimate
Opinionated Go development skill — architecture, conventions, production readiness, repo/CI hygiene, review, and MCP/agent patterns.
kernel-sdks
TypeScript and Python SDK skills for building browser automation with Kernel's Typescript and Python SDKs.
deeplaw-knowledge-os
Explicit-use, read-only access to the local Markdown-native Knowledge OS and bounded Context Capsules.
swift-performance-pro
Performance: Instruments, SwiftUI render cost, memory and retain cycles, launch time, lists.