software-instrumentation

Pass

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill instructions involve the agent writing, reviewing, or modifying software code, which creates a surface for indirect prompt injection if the code being processed contains malicious instructions.
  • Ingestion points: The agent context receives untrusted data in the form of existing source code or requests to generate new code based on user specifications (SKILL.md).
  • Boundary markers: There are no explicit instructions or delimiters defined within the skill to isolate user-provided code or warn the agent to ignore instructions embedded in the code it reviews.
  • Capability inventory: An agent using this skill is expected to have file-writing and code-modification capabilities to implement the recommended instrumentation.
  • Sanitization: The skill does not specify sanitization or validation methods to ensure that comments or strings within the code being instrumented do not influence the agent's behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 22, 2026, 01:16 AM
Security Audit — agent-trust-hub — software-instrumentation