pentest-agents-hunting — Claude Code rule: purpose and setup
The pentest-agents-hunting rule structures security hunting around scope, practical impact, and attack chains. It helps an agent drop weak leads and investigate confirmed bugs more deeply.
no license ★ 1,298 updated 2026-06-12
.github/instructions/pentest-agents-hunting.instructions.mdWhat pentest-agents-hunting does
The rule applies across the project. Before testing, it directs the agent to read the program’s scope limits and judge whether an attacker could cause real harm without unusual actions from a user. A finding with no practical consequence, such as a standalone open redirect, should be set aside.
Automation is reserved for reconnaissance; hands-on testing is used to investigate vulnerabilities. The rule favors understanding one target in depth, checking sibling API endpoints, and looking for related flaws after confirming an initial bug. It also sets limits on unproductive searches: change direction when progress stalls, and investigate a possible second bug before writing up the first.
Scope matters here. An out-of-scope request can put access at risk. Before reporting exposed data, check whether a visitor can already see it through the web interface; for a leaked API key, test what that credential can access.
What it helps with
- discard leads with no practical security impact
- check whether testing stays within the authorized scope
- inspect related API endpoints for access-control flaws
- extend a confirmed bug into a higher-impact attack chain
- verify the permissions granted by a leaked API key
Example prompts for the agent
- Check this finding against pentest-agents-hunting and assess whether it could cause real harm.
- Compare my requests with the testing scope and suggest another endpoint to inspect.
- Look for related endpoints and test whether this vulnerability can form part of an attack chain.
How to install pentest-agents-hunting
Manually: put the files in place
.github/instructions/pentest-agents-hunting.instructions.mdThe applyTo field sets which files it applies to
Manually: put the files in place
.claude/rules/pentest-agents-hunting.mdThe paths frontmatter field limits the files: move applyTo from the source into paths
AGENTS.md
Codex reads AGENTS.md: paste the rule text there
Manually: put the files in place
.cursor/rules/pentest-agents-hunting.mdcFrontmatter: description, globs, alwaysApply. Move applyTo from the source into globs, or the rule will not attach on its own
Manually: put the files in place
.devin/rules/pentest-agents-hunting.mdManually: put the files in place
.clinerules/pentest-agents-hunting.mdStep by step, with commands for each platform: AGENTS.md vs CLAUDE.md: what they are and which one to use
Questions about pentest-agents-hunting
How do I install pentest-agents-hunting in Claude Code?
Add the rule at .claude/rules/pentest-agents-hunting.md.
What does pentest-agents-hunting do?
It guides security testing toward in-scope findings with practical impact, then encourages deeper checks for related bugs and attack chains.
How does the rule handle a possible data leak?
Check in an incognito session whether the same information is visible in the web interface. For a leaked API key, call the API with that credential and inspect what it can access.