Skills
Navigate
skills/chenqg618/compliance-skills/project-billing-collection-check-free/Gen Agent Trust Hub

project-billing-collection-check-free

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The SKILL.md file includes a section '需要完整档时（可选）' containing a text block intended to be interpreted by the AI agent as a direct user command. It uses phrases like 'I have authorized the installation' to persuade the agent to execute a command to install the @alipay/agent-payment tool and initiate a purchase without additional user confirmation steps beyond the payment itself.
  • [INDIRECT_PROMPT_INJECTION]: The skill contains instructional content in its metadata/documentation that could influence the agent's behavior during a session.
  • Ingestion points: SKILL.md (instruction block for the 'Pro' version purchase).
  • Boundary markers: None. The text is provided as an unescaped block for the agent to read.
  • Capability inventory: The skill uses Bash to execute local scripts and Read for file access.
  • Sanitization: None. The agent may execute the suggested npx command if it considers the skill's documentation as a trusted source of user authorization.
  • [COMMAND_EXECUTION]: The skill executes node scripts/run.mjs via the Bash tool. This is the intended local audit functionality and does not involve external network connections.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 11:28 AM
Security Audit — agent-trust-hub — project-billing-collection-check-free