# Tahr Security (ChatGPT Plugin)

> Evidence-backed app security

- Type: plugin
- Publisher: Tahr Security Inc (official)
- Categories: Security
- Website: https://chatgpt.com/plugins/plugins_6aab2358993881919183acd471020907
- Open in ChatGPT: https://chatgpt.com/plugins/plugins_6aab2358993881919183acd471020907
- Version: 0.3.3
- Listed: 2026-10-15

## About

Review applications with an evidence-first security workflow that maps attack surfaces, tests authentication and access controls, traces dangerous inputs, models threats, and verifies fixes.

## Install

1. **Add it in ChatGPT** Open Plugins in the ChatGPT sidebar (web or desktop), search for “Tahr Security” and select +. You'll be asked to sign in to any connected service on first use. Then type @ in a chat to call it.

## Disconnect or uninstall

1. **In ChatGPT** Disconnect an account in Settings → Apps or Plugins. To remove the plugin itself, open Tahr Security in the Plugin Directory and select Uninstall plugin. Uninstalling can disconnect every account connected through it, and plugins your workspace installs may not offer the option.

2. **In Codex** Run /plugins in the Codex CLI, select the plugin and uninstall it, or press Space to turn it off.

## Skills

- **Tahr Audit Android**: Audit Android application security boundaries
- **Tahr Audit Secrets and Config**: Audit secrets, dependencies, and deployment config
- **Tahr Map Attack Surface**: Map the real application attack surface
- **Review Tahr Findings**: Review findings from an existing Tahr account
- **Tahr Secure App**: Run an evidence-backed application security review
- **Tahr Test Access Control**: Prove authorization flaws and reject weak signals
- **Tahr Test AI Agents**: Test LLM, RAG, and agent trust boundaries
- **Tahr Test Authentication**: Test authentication and session boundaries
- **Tahr Test Business Workflows**: Test stateful business logic for abuse
- **Tahr Threat Model App**: Threat-model an entire existing application
- **Tahr Trace Dangerous Inputs**: Trace untrusted input to dangerous sinks
- **Tahr Verify Security Fix**: Verify fixes with exploit-focused regression tests

## Example prompts

- "Run a comprehensive evidence-backed security review of this application and prioritize demonstrated findings."
- "Map this application's attack surface, roles, trust boundaries, routes, parameters, and exposed interfaces."
- "Verify this security fix in the original vulnerable context and test for bypasses and regressions."

## Tahr Security alternatives

- [Codex Security](https://chatgptplugins.directory/plugins/codex-security.md): Security scanning for your codebase
- [Static Analysis (Trail of Bits)](https://chatgptplugins.directory/plugins/trailofbits-static-analysis.md): CodeQL, Semgrep and SARIF parsing skills for security bug hunting
- [Differential Review (Trail of Bits)](https://chatgptplugins.directory/plugins/trailofbits-differential-review.md): Security-focused review of code diffs with blast-radius estimation
- [Supply Chain Risk Auditor (Trail of Bits)](https://chatgptplugins.directory/plugins/trailofbits-supply-chain-risk-auditor.md): Audit npm, PyPI and Go dependencies for supply-chain risk
- [Aivana Security Investigator](https://chatgptplugins.directory/plugins/aivana-security-investigator.md): Run safe KQL investigations
- [Atbash Safety](https://chatgptplugins.directory/plugins/atbash-safety.md): Guard tool calls with Atbash

---
Source: https://chatgptplugins.directory/plugins/tahr-security · Index: https://chatgptplugins.directory/llms.txt · Independent and community-run. Not affiliated with or endorsed by OpenAI. ChatGPT is a trademark of OpenAI.