daily-report
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill leverages local CLI tools including gh, jq, sed, and date to gather and format activity data.
- [EXTERNAL_DOWNLOADS]: The skill executes the acli tool via Nix (nix run nixpkgs#acli), which involves fetching the tool from the Nixpkgs registry.
- [DATA_EXFILTRATION]: The skill processes sensitive data from GitHub, Jira, and Slack. It requires the user to set environment variables for Slack authentication (SLACK_MCP_XOXC_TOKEN and SLACK_MCP_XOXD_TOKEN) to access meeting information from Slack DMs.
- [PROMPT_INJECTION]: The skill processes untrusted data from external platforms (GitHub, Jira, Slack) which could contain malicious instructions meant to influence the agent's behavior.
- Ingestion points: GitHub activity summaries, Jira work items, and Slack message history are fetched via their respective APIs and CLIs.
- Boundary markers: The skill does not implement technical delimiters for the fetched content within its scripts.
- Capability inventory: The skill uses the gh CLI to create issues on GitHub.
- Sanitization: No automated sanitization of fetched text is performed; however, the skill workflow explicitly requires the user to review and edit the draft report before it is published, acting as a manual verification gate.
Audit Metadata