weekly-report-onboard

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill uses uvx to download and execute code directly from a Git repository (github.com/mjkimR/weekly-report@v0.2.0). This repository is owned by the skill author and facilitates the core functionality of the skill.
  • [COMMAND_EXECUTION]: The instructions involve executing several shell commands to probe the environment (command -v uv), manage configuration (weekly-report run --json), and identify repository authors.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8).
  • Ingestion points: The skill reads user-provided content from 'past reports' (Step 3) to infer formatting and tone.
  • Boundary markers: The instructions lack explicit delimiters or instructions to ignore embedded commands within the analyzed reports.
  • Capability inventory: The agent has the ability to execute shell commands via uvx, read local Git repositories, and write configuration files to ~/.weekly-report/.
  • Sanitization: There is no mention of sanitizing or filtering the content of the reports before processing them for format extraction.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 11:30 PM
Security Audit — agent-trust-hub — weekly-report-onboard