Pentest Copilot is an AI-driven penetration-testing agent that connects to a Kali attack box, runs security tools, interprets their output, and chooses subsequent actions. It is intended for ethical hacking engagements, boot-to-root challenges, and capture-the-flag competitions. Catalogue add-ons extend its workflows.
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/bugbasesecurity/pentest-copilotWrote 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/rules/bugbasesecurity/pentest-copilot/shell-tilde-paths)<a href="https://agentmods.dev/rules/bugbasesecurity/pentest-copilot/shell-tilde-paths"><img src="https://agentmods.dev/badge/rules/bugbasesecurity/pentest-copilot/shell-tilde-paths/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/rules/bugbasesecurity/pentest-copilot/shell-tilde-paths"><img src="https://agentmods.dev/badge/rules/bugbasesecurity/pentest-copilot/shell-tilde-paths.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00258 |
| Opus 5.5 | $0.00000 | $0.00103 |
| Sonnet 5.5 | $0.00000 | $0.00052 |
| Haiku 4.5 | $0.00000 | $0.00026 |
Grade A, and why
shell-tilde-paths scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Shell Path Handling: ~ Does Not Expand in Quotes
WORKSPACE_DIR is ~/pentest-workspace. When constructing shell commands that reference file paths:
~is not expanded by bash inside double quotes:stat "~/foo"fails (looks for literal~directory)$HOMEis expanded inside double quotes:stat "$HOME/foo"works correctly
Always replace a leading ~ with $HOME before interpolating paths into quoted shell commands:
// ✅ GOOD
const resolved = rawPath.replace(/^~(?=\/|$)/, "$HOME");
await ctx.runCommand(`stat -c %s "${resolved}"`);
// ❌ BAD — ~ won't expand inside double quotes
await ctx.runCommand(`stat -c %s "${rawPath}"`);
// ❌ BAD — unquoted path breaks on spaces
await ctx.runCommand(`stat -c %s ${rawPath}`);
This applies to any tool handler that takes a user/agent-supplied file path and passes it to ctx.runCommand.
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.
- 1mo ago First seen · 29 lines · 0 tokens per session scan A 44e04dbeb0a4
shell-tilde-paths is a cursor rule published in the GitHub repository bugbasesecurity/pentest-copilot (1,537 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 258 tokens. 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 2026-08-30.
Other cursor rules, from other repositories
no-jumping-ahead
ALWAYS follow these rules when working on this codebase.
agent-autonomy
AI agent autonomy threat agent — detects excessive autonomy, goal misalignment, unconstrained action scope, missing human-in-the-loop checkpoints, cascading multi-agent failures, and autonomous resource consumption against Processes.
ipynb-files
This file contains rules and prompts for handling Jupyter Notebook (.ipynb) files in Cursor IDE, which currently doesn't have native support for .ipynb files.
safari-automation
Patterns and best practices for automating Safari browser interactions for web UI automation and testing.
dependencies
Dependency management guidelines.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.