
open-sesame
π An AI web-pentest skill for Claude Code. Point it at a site you're authorized to test β it maps the attack surface, works through the OWASP WSTG, validateβ¦
Install with your AI
Paste into Claude Code, Cursor, or any agent β it reads the repo and wires the tool into your project.
Install and set up open-sesame (release-binary project) into my current project. Found on https://claudeers.com/open-sesame Repo: https://github.com/hassan-algo/open-sesame Homepage/docs: β Detected install method: release-binary β inspect the README Category: security. Platforms: cli, api, web. Read the repo's README for exact setup and env vars, then install it and wire it into my project. Claudeers Health Verdict: active; community-verified: false. Confirm the source before running anything.
Grab the latest release asset from GitHub.
# download a build from https://github.com/hassan-algo/open-sesame/releases
git clone https://github.com/hassan-algo/open-sesame
// compatibility
| Platforms | cli, api, web |
|---|---|
| Operating systems | β |
| AI compatibility | claude |
| License | NOASSERTION |
| Pricing | open-source |
| Language | HTML |
open-sesame π
An AI web-application penetration-testing skill for Claude.
Say the magic words, and the web app opens up β the authorized way.
open-sesame is a skill that turns Claude into a disciplined web-app penetration
tester. Give it a target you're authorized to assess and it runs the whole engagement β
maps the attack surface, works methodically through the OWASP Web Security Testing
Guide, validates findings
with the lightest safe proof, and writes clean reports for both engineers and
leadership.
It's built as a set of Markdown instructions and small helper scripts β no model fine-tuning, no binaries. You can read every line of what it tells Claude to do.
[!WARNING] Authorized testing only. This skill is for systems you own or have explicit permission to test (your own apps, a signed client engagement, an in-scope bug-bounty target, or a CTF / practice lab). It is designed to refuse unauthorized targets, treat the platform/host a target runs on as out of scope by default, and prefer non-destructive proofs. Using it against systems you don't have permission to test may be illegal. You are responsible for staying within scope and the law.
β¨ What it does
- Authorization-first β confirms scope and rules of engagement before any active test.
- Full OWASP WSTG coverage β recon, configuration, authentication, session management, access control (incl. multi-tenant isolation), injection, error/crypto, business logic, client-side, and API-specific tests.
- Tooling playbook β how to drive Burp Suite, ffuf, nuclei, sqlmap, nmap, subfinder/httpx and friends, chained into a repeatable workflow.
- Authenticated & multi-tenant testing β what to request from the client and how to prove cross-user / cross-tenant access safely.
- Safe RCE validation β prove code execution with the lightest signal first; reverse/bind shells, TTY upgrade, and strict cleanup rules only when impact must be shown.
- Reports for two audiences β a technical findings report and a plain-language stakeholder summary, each rendered to PDF.
π Quick start
1. Install the skill
Clone into a location Claude discovers skills from:
# Per project (available inside one repo)
git clone https://github.com/hassan-algo/open-sesame.git
cp -r open-sesame/.claude/skills/web-pentest .claude/skills/
# Or personal (available in every project)
cp -r open-sesame/.claude/skills/web-pentest ~/.claude/skills/
2. Use it
In Claude Code, just describe the work β the skill triggers automatically:
Pentest my web app at https://staging.myapp.example β I own it. Start with recon.
It also triggers on phrases like "find vulns in my web app", "test this endpoint for
IDOR/SQLi", "map the attack surface", "write up this finding", or "validate this
RCE" β or invoke it explicitly with /web-pentest.
3. Get PDF reports
After a report is drafted in Markdown, render it to a styled PDF:
.claude/skills/web-pentest/scripts/render-report.sh report.md report.pdf
The script converts Markdown β HTML (pandoc, else npx marked) β PDF (headless
Chrome/Chromium, else weasyprint, else macOS cupsfilter). Install at least one tool
from each stage.
ποΈ How it's organized
.claude/skills/web-pentest/
βββ SKILL.md # entry point: authorization, workflow, pointers
βββ references/
β βββ methodology.md # the OWASP WSTG test checklist (the core)
β βββ tooling.md # tool commands & safe-usage workflows
β βββ post-exploitation.md # RCE validation, reverse shells, cleanup
β βββ reporting.md # severity (CVSS), evidence, dual-audience + PDF
β βββ lab-targets.md # legal practice targets & a learning path
βββ assets/
β βββ report-template.md # fill-in findings report
βββ scripts/
β βββ render-report.sh # Markdown report β styled PDF
βββ evals/
βββ evals.json # trigger/behavior test cases
βββ benchmarks/ # validation results
Skills use progressive disclosure: SKILL.md is always loaded; the references/
files are pulled in only when a phase needs them, keeping the context lean and the
workflow fast.
π§ͺ Validation
The skill was built and tuned with Anthropic's skill-creator and validated across two
eval iterations β methodology, reporting, authorization-gating, RCE/reverse-shell
guidance, cross-tenant IDOR, and scope discipline. Summary results live in
.claude/skills/web-pentest/evals/benchmarks/.
π€ Contributing
Contributions are very welcome β this gets better the more practitioners shape it. Good first contributions:
- Extend the methodology β add or sharpen WSTG test cases in
references/methodology.md. - Add tooling recipes β real-world command workflows in
references/tooling.md. - Improve reporting β templates, severity guidance, or the
render-report.shbackends. - Add eval cases β new prompts in
evals/that catch gaps or regressions.
See CONTRIBUTING.md for how to propose changes. Please keep everything aligned with the skill's core principle: authorized, non-destructive, methodical testing β contributions that add offensive capability without that framing won't be merged.
π License
MIT β free to use, modify, and share.
π Acknowledgements
- Methodology based on the OWASP Web Security Testing Guide and the OWASP Top 10.
- Built as a Claude Agent Skill.
// faq
What is open-sesame?
π An AI web-pentest skill for Claude Code. Point it at a site you're authorized to test β it maps the attack surface, works through the OWASP WSTG, validates findings, and writes dev + stakeholder reports (PDF).. It is open-source on GitHub.
Is open-sesame free to use?
open-sesame is open-source under the NOASSERTION license, so it is free to use.
What category does open-sesame belong to?
open-sesame is listed under security in the Claudeers registry of Claude-compatible tools.
// embed badge
[](https://claudeers.com/open-sesame)
// retro hit counter
[](https://claudeers.com/open-sesame)
// reviews
// guestbook
// related in Security & Compliance
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a speciaβ¦
Ο RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection β all without a single pixel of video.
Prowler is the worldβs most widely used open-source cloud security platform that automates security and compliance across any cloud environment.
πΆ A curated list of Web Security materials and resources.