context-compressor
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data types such as conversation history, code files, and research results (SKILL.md).\n
- Ingestion points: Processes external data from various sources as outlined in Step 1 of the Protocol (SKILL.md).\n
- Boundary markers: The instructions do not define clear delimiters or directives to ignore instructions that might be embedded within the processed text.\n
- Capability inventory: The skill possesses no capabilities for network access, file modification, or shell command execution, which prevents the exploitation of any potential injection.\n
- Sanitization: No input sanitization or filtering is specified for the content before it is summarized.\n- [NO_CODE]: The skill consists exclusively of markdown instructions and YAML configuration files, with no executable scripts, binaries, or remote package dependencies identified.
Audit Metadata