x-twitter-growth
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONMETADATA_POISONING
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill scripts ingest untrusted external data from Twitter profiles without proper sanitization.
- Ingestion points: Profile handles and bios are passed as command-line arguments to
scripts/profile_auditor.pyandscripts/competitor_analyzer.py. - Boundary markers: The skill does not use delimiters or instructions to ignore embedded commands in the ingested data.
- Capability inventory: The skill can write to the local file system through
scripts/growth_tracker.py. - Sanitization: The scripts lack sanitization for directory traversal characters in handles, which can lead to file writes in unintended locations within the skill directory.
- [METADATA_POISONING]: The metadata in
SKILL.mdlists a future update date of 2026-03-10, which is inaccurate.
Audit Metadata