fastmail
Warn
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: MEDIUMDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill reads sensitive credentials from a configuration file located at
~/.config/fastmail/creds.\n - Evidence: Both
scripts/calendar.pyandscripts/send.pyuse functions (get_credentialsandload_credsrespectively) to extract a user's email and app password from this file for authentication.\n - Context: Access to this file is required for the skill's primary function of interacting with Fastmail's SMTP and CalDAV servers.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted data which is subsequently used in network operations.\n
- Ingestion points: Data is ingested through parameters such as
--bodyinscripts/send.pyand--title,--description, and--locationinscripts/calendar.py.\n - Boundary markers: No boundary markers or instructions are used to separate untrusted data from the system logic.\n
- Capability inventory: The skill possesses the ability to read local filesystem data and perform network requests to Fastmail infrastructure.\n
- Sanitization: Input data is processed without escaping or validation, allowing potentially malicious instructions to be included in email bodies or calendar entries.
Audit Metadata