language-audit
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (websites and HTML source) while possessing capabilities that could be abused if the agent follows instructions hidden in that data.\n
- Ingestion points: The skill ingests content from external URLs, sitemaps, and HTML source code provided for auditing (SKILL.md).\n
- Boundary markers: There are no instructions to use delimiters or ignore embedded commands when processing the external content.\n
- Capability inventory: The skill reads brand configuration files from
~/.claude-marketing/and executes a local command-line tool,language-router.py, using content sampled from the external sources (SKILL.md, Process steps 1 & 5).\n - Sanitization: No sanitization or escaping of the ingested data is specified before it is processed or used in tool calls.\n- [COMMAND_EXECUTION]: The skill uses a command-line tool to process data extracted from external websites.\n
- Evidence: Process step 5 executes
language-router.py --action scoreusing sampled content from the audited website (SKILL.md).
Audit Metadata