The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill has a surface where untrusted data from user files is ingested and presented to the agent, potentially allowing embedded instructions to influence the agent's behavior.\n
Ingestion points: The script scripts/i18n_checker.py reads project files using the pathlib module's read_text method.\n
Boundary markers: There are no explicit delimiters or safety instructions used to wrap or contextually isolate the findings reported by the checker script.\n
Capability inventory: The skill is granted Read, Glob, and Grep capabilities. The Python script performs recursive file discovery and reads content to identify text patterns.\n
Sanitization: The tool does not perform sanitization, escaping, or filtering of the strings it extracts from scanned files before displaying them to the agent.

i18n-localization