Security Investigation
Independent PiSkill directory guide. The original prompt remains hosted by Microsoft PromptKit.
What does this prompt do?
Investigates a suspected security weakness by tracing trust boundaries, inputs, sensitive operations, exploitability, affected assets, and practical remediation evidence.
Primary use case
Assess a concrete suspected vulnerability in code or architecture.
Expected output
A scoped exploitability assessment with evidence and recommended fixes.
Inputs or variables
- suspected issue
- code
- security context
Related prompts
Incident Response Plan
Structures an incident response workflow around evidence collection, containment, recovery and follow-up actions.
LLM Security Review
Reviews an LLM or agent solution for prompt injection, unsafe tool access, data exposure, trust-boundary failures, weak validation, and operational controls.
Responsible AI Evidence Capture
Captures responsible-AI requirements, stakeholders, harms, controls, evidence, ownership, and unresolved questions in a consistent planning artifact.
Responsible AI Plan from PRD
Turns a product requirements document into a responsible-AI plan covering likely harms, mitigations, validation, monitoring, governance, and accountable owners.
Responsible AI Plan from Security Plan
Extends an existing security plan with model-specific harms, misuse, data, evaluation, transparency, monitoring, and governance requirements.
Secure Supply Chain Plan from PRD
Derives software-supply-chain controls from product requirements, covering dependencies, provenance, builds, artifacts, releases, ownership, and verification.