daily-chief
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses
npxto download and execute thedaily-chief-agent@1.0.0package from the NPM registry. While the version is pinned, this introduces a dependency on external code hosted on a public registry. - [COMMAND_EXECUTION]: The skill invokes multiple shell commands via
npx, including authentication and planning workflows. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) by processing user inputs into command-line arguments.
- Ingestion points: Step 2 in
SKILL.mdinstructs the agent to gather answers from the user conversation (available hours, energy, tasks). - Boundary markers: Absent. The instructions do not provide delimiters or specific guidelines for the agent to separate untrusted user data from the command structure.
- Capability inventory: The skill utilizes shell command execution via
npxand requires network access to a remote server. - Sanitization: Absent. The instructions in Step 3 interpolate user answers directly into a JSON string within a shell command (
--answers-json). There are no instructions to escape shell-sensitive characters or validate the structure, creating a risk of command injection if the agent naively interpolates malicious user input.
Audit Metadata