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 zanwei/harness-designnpx agentmods add plugins/zanwei/harness-design/marketplacegit clone --depth 1 https://github.com/zanwei/harness-designGrade A, and why
harness-design 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 2d 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": "harness-design",
"owner": {
"name": "Zanwei Guo",
"email": ""
},
"metadata": {
"description": "Browser-based verification harness for design-dna Phase 3 output (HTML vs Design DNA JSON + Reference URL)",
"version": "2.0.0"
},
"plugins": [
{
"name": "harness-design",
"description": "Verify, score, and correct generated designs against Design DNA JSON + reference URL with browser-based visual verification",
"source": "./",
"strict": true,
"skills": ["./skills/harness-design"]
}
]
}
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.
- 2d ago First seen · 21 lines scan A a034471c99ac
harness-design marketplace is a plugin published in the GitHub repository zanwei/harness-design (4 stars, last pushed 4mo 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
data-analysis-ml-agent-skills
Evidence-first AI agent skills for reliable data analysis, machine learning, and reproducible model validation.
actions
GitHub Actions skills — least-privilege workflow style with truthy/falsey expressions, security hardening (SHA-pinning, no pullrequesttarget), wiring the bitwise reusable-workflow library, and an actionlint/zizmor validation loop.
golang
Modern Go development — stdlib-first code style with cobra/viper CLIs, table-driven tests and native fuzzing, canonical project layout with pinned tooling, documentation conventions with LLM-ready CLI references, and GoReleaser-based release engineering.
python
Modern Python on the Astral toolchain — uv for project and dependency management with the uvbuild backend, ruff for formatting and linting, ty or pyright for type checking, pytest with Hypothesis property tests, Google-style docstrings with CLI reference generation, and trusted-publishing release engineering.
workflow
Developer-workflow commands and skills — sandbox-safe Conventional commits with a commit.sh handoff, immutable code-scanning triage reports with SHA-pinned permalinks, and generation of evolve evaluation suites (triggers + behavioral evals) for agent skills.
terraform
Terraform style conventions, module scaffolding, and fmt/validate/tflint workflows for authoring reusable modules.