meta-ads-open-cli

Warn

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: MEDIUMDATA_EXFILTRATIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill accesses sensitive authentication data through the META_ADS_ACCESS_TOKEN environment variable and a configuration file located at ~/.config/meta-ads-open-cli/credentials.json.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the global installation of the meta-ads-open-cli package from the npm registry, which is a tool provided by the author.
  • [COMMAND_EXECUTION]: Multiple shell commands are executed using the meta-ads-open-cli tool to query Meta's marketing platform for campaign stats and lead information.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its handling of untrusted data from external marketing sources. 1. Ingestion points: Data enters the context via commands like leads (form submissions) and creative (ad content). 2. Boundary markers: There are no boundary markers or instructions to the agent to disregard potential commands embedded in the marketing data. 3. Capability inventory: The agent has the ability to run various shell-based data retrieval operations. 4. Sanitization: The skill lacks logic to sanitize or validate the content retrieved from Meta's API endpoints.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 9, 2026, 06:31 PM