The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize external brand mentions from news, reviews, or social media, which are untrusted data sources.
Ingestion points: The run.py script reads JSON payloads from a file path provided via the --input command-line argument. SKILL.md defines the input as JSON objects or CSV rows containing business signals.
Capability inventory: The provided Python script has the capability to read from and write to the local file system using pathlib.Path. While no network or subprocess execution is currently present in the scaffold, the output is intended for agent processing.
Boundary markers: There are no delimiters or instructions in SKILL.md to prevent the agent from following malicious commands that might be embedded in the brand mention data.
Sanitization: There is no evidence of input validation, filtering, or sanitization in the run.py implementation.

brand-mention-sentiment-tracker