Claude Vulnerability Detection Agent
Independent PiSkill directory guide. The original prompt remains hosted by Anthropic Claude Cookbooks.
What does this prompt do?
Builds an agent that inspects code for vulnerability evidence, uses tools to gather context, reports uncertainty, and produces actionable remediation rather than unsupported alarms.
Primary use case
Investigate code for security vulnerabilities with an agent.
Expected output
Evidence-backed vulnerability findings and remediation steps.
Inputs or variables
- codebase
- security scope
- tool access
Related prompts
One-Line Research Agent
Demonstrates a compact Claude Agent SDK research workflow that delegates source gathering, preserves citations, and synthesizes results around a bounded question.
Iterative Repair Loops with Codex
Shows how to structure an iterative coding loop that observes failures, changes a bounded implementation, reruns checks, and stops on success or a defined limit.
Exhaustive Bug Hunt
Runs a systematic bug investigation across changed behavior, edge cases, error paths, concurrency, state, and integration boundaries with evidence requirements.
Claude Citation Workflow
Shows how to request, preserve, and validate citations so claims remain traceable to supplied source material and unsupported statements can be identified.
Claude Content Moderation
Creates a Claude moderation workflow from explicit policy categories, severity rules, examples, structured decisions, uncertainty handling, and evaluation data.
Claude Evaluation Builder
Builds an evaluation set and scoring workflow for a model task using representative inputs, clear criteria, repeatable execution, and analysis of failure patterns.