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 William2333ZZ/trustshellnpx agentmods add plugins/william2333zz/trustshell/marketplacegit clone --depth 1 https://github.com/William2333ZZ/trustshellGrade A, and why
trustshell 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": "trustshell",
"owner": {
"name": "TrustShell",
"email": "[email protected]"
},
"metadata": {
"description": "AI agent security for Claude Code — read the code, break the running agent, prove it (exploit-validated)."
},
"plugins": [
{
"name": "trustshell",
"source": "./",
"description": "Red-team AI agents from inside Claude Code: /static-scan triages an agent's source for candidate vulnerable paths, /red-team attacks a running agent and CONFIRMS what's real with a proof marker, and the agent-red-teamer subagent runs the whole exploit-validated loop. Ships the RT-1..RT-9 attack skills.",
"author": {
"name": "TrustShell"
},
"homepage": "https://trust-shell.com",
"repository": "https://github.com/William2333ZZ/trustshell",
"license": "MIT",
"keywords": [
"ai-agent-security",
"red-team",
"prompt-injection",
"memory-poisoning",
"security",
"exploit-validated",
"agent-security"
],
"category": "security",
"tags": [
"ai-agent-security",
"red-team",
"prompt-injection",
"memory-poisoning",
"security",
"exploit-validated"
]
}
]
}
What ships with it
1 file beside marketplace.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.
- yesterday First seen · 42 lines scan A 2884737d5168
trustshell is a plugin published in the GitHub repository William2333ZZ/trustshell (1 stars, last pushed 1mo 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
emilia-protocol
Plugin marketplace listing 1 plugin: emilia-guard.
claude-code-plugin
Install and manage the local-first HOL Guard package for Claude Code. This plugin guides setup and status checks; HOL Guard itself installs and owns runtime security hooks for risky commands, secrets, packages, MCP, and supply-chain activity.
claude-plugin-good
Example Claude plugin package.
marketplace
Plugin marketplace listing 1 plugin: praxen.
peephole
Plugin marketplace listing 1 plugin: peephole.
peephole
A plugin for coding agents. They ship secure code. Steers the write and can ask or deny before a file is saved.