write-migration

Pass

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill instructs the agent to read and modify codebase content, which presents a standard surface for indirect prompt injection. Malicious instructions embedded in the source files could potentially influence the agent's task execution.
  • Ingestion points: The agent processes source code, registry lookups, and test fixtures during migration waves (documented in SKILL.md rules 3 and 5).
  • Boundary markers: No specific delimiters are mandated to separate ingested code from the operational prompt.
  • Capability inventory: The agent can write to files and execute shell commands defined in the local repository's AGENTS.md configuration.
  • Sanitization: The guide does not include requirements to sanitize or validate code content prior to processing.
  • [SAFE]: The skill enforces safety-conscious development workflows, such as prohibiting self-review and requiring proof of equivalence for all changes. No obfuscated strings, credential theft patterns, or unauthorized remote code executions were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 06:03 PM
Security Audit — agent-trust-hub — write-migration