Skills
skills/alekseiul/sprut-agent-kit/analytics/Gen Agent Trust Hub

analytics

Warn

Audited by Gen Agent Trust Hub on Apr 16, 2026

Risk Level: MEDIUMCREDENTIALS_UNSAFEREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill accesses sensitive Google OAuth2 client secrets from a configuration path outside of its own directory. Evidence: In scripts/yt-analytics-auth.py, the script reads from ~/.config/gog/client_secret.json to facilitate authentication.
  • [REMOTE_CODE_EXECUTION]: The skill utilizes the pickle module for deserializing local authentication tokens, which is an insecure practice that can lead to code execution if the token file is modified. Evidence: pickle.load() is used in scripts/yt-analytics-auth.py and scripts/yt-deep-stats.py to read ~/.openclaw/yt-analytics-token.pickle.
  • [PROMPT_INJECTION]: The skill ingests untrusted data from Telegram messages and YouTube metadata, which could influence agent behavior. Ingestion points: Telegram message retrieval in scripts/tg-stats.py and YouTube title retrieval in scripts/yt-deep-stats.py. Boundary markers: Absent. Capability inventory: Environment variable access and shell command execution. Sanitization: Limited to truncation and newline removal.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 16, 2026, 05:29 PM
Security Audit — agent-trust-hub — analytics