review
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEOBFUSCATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [OBFUSCATION]: The persona reference file
references/stb-reviewer.mdcontains zero-width space characters (U+200B) within the suggested signature strings (e.g., 'Amy\u200B'). The file's documentation explicitly notes that these are included for 'authenticity' in the persona's signature. While documented, zero-width characters are a standard obfuscation technique used to hide data from users or bypass text-matching filters. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted code and diffs provided by users. It does not mandate the use of boundary markers (such as XML tags) or specific instructions to ignore embedded commands within the code being reviewed. This represents an attack surface for indirect prompt injection, where malicious instructions in a code sample could influence the agent's behavior.
- Ingestion points: Code and diff context provided to the agent as described in the
SKILL.mdworkflow. - Boundary markers: Absent; the instructions do not require the agent to wrap untrusted data in specific delimiters.
- Capability inventory: The skill uses
Read,Grep,Glob, andBashtools to inspect project source code and history. - Sanitization: Absent; no sanitization or validation of the user-provided code context is performed before it is processed by the persona-driven review logic.
Audit Metadata