terraform-plan-risk-auditor
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external sources, specifically Terraform plan JSON files, via the plan_json, plan_json_text, or plan_path parameters.
- Ingestion points: Data enters the execution context through the plan_json, plan_json_text, or plan_path parameters processed in scripts/main.py.
- Boundary markers: The application relies on standard JSON parsing as a boundary but does not provide explicit instructions or delimiters to prevent the agent from acting on instructions that might be embedded within the data values of the resource changes.
- Capability inventory: The skill is restricted to local file reading and outputting structured text to standard output; it does not have network access or administrative execution capabilities.
- Sanitization: While the script validates that the input is well-formed JSON, it does not sanitize or escape the content of string values before they are used in heuristic logic and reported to the agent.
Audit Metadata