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 pr3m/claude-code-bash-smart-approvenpx agentmods add plugins/pr3m/claude-code-bash-smart-approve/bash-smart-approvegit clone --depth 1 https://github.com/pr3m/claude-code-bash-smart-approveGrade A, and why
bash-smart-approve 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": "bash-smart-approve",
"version": "0.2.0",
"description": "Auto-approves Claude Code Bash commands via AST-based allowlist. Handles pipes, subshells, compound commands, and env-var prefixes that literal Bash(...) prefix rules can't match.",
"author": {
"name": "Christjan Schumann",
"url": "https://github.com/pr3m"
},
"homepage": "https://github.com/pr3m/claude-code-bash-smart-approve",
"repository": "https://github.com/pr3m/claude-code-bash-smart-approve",
"license": "MIT",
"keywords": ["claude-code", "plugin", "hook", "bash", "permissions", "allowlist", "pretooluse"]
}
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 · 14 lines scan A f4e70ec5b653
bash-smart-approve is a plugin published in the GitHub repository pr3m/claude-code-bash-smart-approve (1 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
agmsg
Cross-agent messaging via SQLite. Send messages between CLI AI agents. No daemon, no network.
caushell-claude
Session-aware shell guardrails for Claude Code backed by the Caushell runtime.
agent-messaging
Agent Messaging Protocol (AMP) - Secure, cryptographically signed messaging between AI agents across different providers.
branch-guard
Cut routine git/gh approval prompts: auto-approve safe commands (read-only, staging, branch/worktree create, feature-branch commits & pushes) and confirm protected-branch or destructive ones.
prompt-improver
Turn vague agent prompts into precise, verifiable XML specs — then execute or plan. Headless multi-CLI generator. Use when the user says improve prompt, plan carefully, or structure a vague coding request.
simple-statusline
A simple, hackable two-line statusline for Claude Code showing model, git status, context usage, and rate limits.