codeguard-erlang
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes untrustworthy Erlang source code and project configurations, which constitutes a vulnerability surface for indirect prompt injection attacks where malicious instructions could be embedded in data.
- Ingestion points: As described in SKILL.md, the agent reads Erlang source files (.erl, .hrl) and project configuration files (rebar.config, elvis.config).
- Boundary markers: The instructions do not prescribe the use of clear delimiters or instructions to ignore embedded commands within the ingested files.
- Capability inventory: The skill utilizes local shell execution for tools like elvis and erlfmt and maintains the capability to suggest or apply file modifications.
- Sanitization: There are no defined procedures for sanitizing or escaping the content of project files before they are incorporated into the agent's reasoning context.
Audit Metadata