stakeholder-summary
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- Indirect Prompt Injection Surface: The skill is designed to process external documents, specifically legal contract reviews and matter-specific files. Because these sources can contain untrusted content, there is a potential for instructions embedded within the text to influence the agent's summarization behavior.
- Ingestion points: The skill reads contract review memos and files located in
~/.claude/plugins/config/claude-for-legal/commercial-legal/matters/. - Boundary markers: The instructions do not define specific delimiters or "ignore" directives to isolate external content from the core instructions.
- Capability inventory: The skill instructs the agent to read existing configuration and matter files and to write outputs to specific directories. It does not utilize network access or shell command execution.
- Sanitization: There are no explicit instructions to sanitize or validate the content of the external summaries or the user-provided matter identifiers before processing them.
- File Path Management: The skill organizes its output using a "matter-slug" provided by the user to determine the directory path. While this is used for legitimate document organization, it introduces a consideration for how identifiers are handled when constructing local file paths.
Audit Metadata