sred-work-summary
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
ghCLI and standard shell commands (date,find,sed) to identify repositories and dates. These operations are limited to the user's local environment and specific GitHub organizations. - [DATA_EXPOSURE]: The skill aggregates potentially sensitive work data (PRs, documents, tickets) from GitHub, Linear, and Notion. However, it explicitly directs the agent to place this data into a private Notion document for the user's own reporting purposes, following standard workflow patterns.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data (PR descriptions, Notion document content, and Linear tickets). While this presents a surface for indirect prompt injection, the instructions are focused on grouping and summarizing rather than executing instructions found within that data. Sanitization is handled by the agent's internal reasoning during the grouping process.
Audit Metadata