contracts-renewals
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists exclusively of markdown documentation and YAML frontmatter. It contains no executable scripts, shell commands, or software code.- [PROMPT_INJECTION]: The skill identifies a workflow for indirect prompt injection. The agent is instructed to ingest and analyze external, untrusted data sources (third-party contracts and MSAs). This could allow malicious actors to embed instructions within those documents to manipulate the agent's output.
- Ingestion points: The agent ingests signed contracts from platforms like DocuSign and PandaDoc (SKILL.md).
- Boundary markers: No delimiters or 'ignore' instructions are specified to isolate document content.
- Capability inventory: The agent is tasked with extracting metadata, indexing terms, and generating renewal quotes based on contract data (SKILL.md).
- Sanitization: No sanitization or validation steps for ingested contract content are described.
Audit Metadata