sred-work-summary

Pass

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes local shell commands to automate the discovery of work items and environmental context.
  • Evidence: Uses date +%Y to calculate the reporting window and find . -maxdepth 2 -name ".git" -type d | sed 's/\/.git$//' | sort to identify local developer repositories for analysis.
  • [DATA_EXFILTRATION]: The skill aggregates data across GitHub, Linear, and Notion to create a centralized report.
  • Evidence: Fetches PR descriptions, ticket titles, and document content via MCP tools and the gh CLI. The resulting summary is stored in a private Notion document within the user's workspace, which is a well-known service.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes and summarizes untrusted external data.
  • Ingestion points: Step 6 processes full content from Notion documents and descriptions from GitHub PRs and Linear tickets.
  • Boundary markers: None present; there are no instructions to the agent to disregard commands embedded within the fetched content.
  • Capability inventory: The skill has read/write access to Notion and Linear via MCPs and GitHub via the gh CLI (SKILL.md).
  • Sanitization: No sanitization or filtering is performed on the ingested content before it is used for summarization.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 28, 2026, 12:35 AM