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/PBNZ/pbnz-skillsWrote 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/pbnz/pbnz-skills/marketplace)<a href="https://agentmods.dev/plugins/pbnz/pbnz-skills/marketplace"><img src="https://agentmods.dev/badge/plugins/pbnz/pbnz-skills/marketplace/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/pbnz/pbnz-skills/marketplace"><img src="https://agentmods.dev/badge/plugins/pbnz/pbnz-skills/marketplace.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
pbnz-skills marketplace 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 9d 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
{
"$schema": "https://anthropic.com/claude-code/marketplace.schema.json",
"name": "pbnz-skills",
"owner": {
"name": "PBNZ"
},
"metadata": {
"description": "A personal collection of Agent Skills for Claude Code, Claude.ai, the Claude API, and other Agent-Skills-compatible tools.",
"version": "0.6.2"
},
"plugins": [
{
"name": "watch-local",
"source": {
"source": "git-subdir",
"url": "https://github.com/PBNZ/watch-local.git",
"path": "plugins/watch-local"
},
"description": "Give Claude a video input — fully local. /watch downloads, extracts frames, and transcribes on a self-provisioned portable runtime (no Docker, no system installs; delete one folder to remove everything) — NVDEC decode + CUDA whisper on a detected NVIDIA GPU, CPU-only mode otherwise. No cloud APIs. Windows 11 primary; Linux/macOS via pwsh (CPU mode, less tested). Canonical home: PBNZ/watch-local.",
"category": "productivity"
},
{
"name": "repokit",
"source": {
"source": "git-subdir",
"url": "https://github.com/PBNZ/repo-kit.git",
"path": "plugins/repokit"
},
"description": "Scaffold standard-compliant repos (/new-repo) and apply the RepoKit repo standard while you work. Canonical home: PBNZ/repo-kit.",
"category": "productivity"
},
{
"name": "newton",
"source": {
"source": "git-subdir",
"url": "https://github.com/PBNZ/newton-skill.git",
"path": "plugins/newton"
},
"description": "Reasoning and sparring partner mode for rigorous, calibrated engagement — honest pushback, current-sources research, reuse-before-rebuild checks, and surgical edits. Canonical home: PBNZ/newton-skill.",
"category": "productivity"
}
]
}
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.
- 9d ago First seen · 44 lines scan A d1eaa954937a
pbnz-skills marketplace is a plugin published in the GitHub repository PBNZ/pbnz-skills (1 stars, last pushed 25d ago), licensed Apache-2.0. 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
bonsai
Silent observer that surfaces high-signal observations between turns.
vibe-ship
Generates a complete production-ready deployment setup for any app in one pass: Dockerfile, docker-compose, CI/CD pipeline, security hardening, and scalability config. Also audits existing Dockerfiles and CI setups with a static scoring script.
swarm-skill
Multi-agent codebase audit skill for Claude Code — spawns parallel subagents (security, performance, tests, architecture, dead-code) and synthesizes their findings into a prioritized action plan.
houssem-plugins marketplace
Claude Code plugins by Houssem Djeghri.
bridger
Let two Claude Code sessions talk to each other: named peers, file-based message threads, ask/answer with full live context.
dev-pipeline
A 5-phase development pipeline (requirement engineering, system architecture, task generation, verified implementation, code review) with an optional QA gate (plan-qa + execute-qa) for changes that have a running surface worth driving — independent of /review, run sequentially or in parallel at the developer's…