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 Ronkupper/PRISMnpx agentmods add plugins/ronkupper/prism/marketplacegit clone --depth 1 https://github.com/Ronkupper/PRISMGrade A, and why
prism-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 yesterday.
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": "prism-marketplace",
"owner": { "name": "Ron Kuper", "url": "https://github.com/Ronkupper" },
"metadata": {
"description": "PRISM framework, distributed as a Claude Skill plugin.",
"version": "1.1.0"
},
"plugins": [
{
"name": "prism",
"version": "2.21.0",
"source": "./plugins/prism",
"description": "PRISM — structured, multi-session, multi-vendor LLM-orchestrated audit & research framework. Loads a lean core and fetches reference bundles (Lens Library, templates, appendices) on demand."
}
]
}
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.
- yesterday First seen · 17 lines scan A 6f8c73881ade
prism-marketplace is a plugin published in the GitHub repository Ronkupper/PRISM (5 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
starknet-agentic-skills
Plugin marketplace listing 1 plugin: starknet-agentic-skills.
ai-news-hq
An AI-news crew (scout, analyst, editor) that tracks the field via web search and ships a ranked briefing on a cadence.
swarmgpt
Plugin marketplace listing 1 plugin: swarmgpt.
researcher-ai
Auditable and sandbox-aware autonomous research workflows.
hackingtool
183+ pentesting and OSINT tools (nmap, nuclei, amass, subfinder, httpx, sherlock, maigret, trufflehog, sqlmap, impacket, and more) wired into Claude Code as a single skill. Runs locally on any OS via native Bash, WSL, or purpose-built Docker images (instrumentisto/nmap, projectdiscovery/nuclei, caffix/amass, etc.).…
jury
Jury v0.1.0 — standalone cross-family cascade-then-jury reviewer (Gemini + GPT small-fish → cross-family tiebreaker). The free 'jury beats judge' base; codi and super-developer depend on it optionally for cross-family verification. No API keys (OAuth local CLIs).