The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external business signals from JSON or CSV files, which creates a surface for indirect prompt injection if the processed data is later used in LLM prompts.
Ingestion points: run.py reads content from a user-specified file path using input_path.read_text() and parses it as JSON.
Boundary markers: The implementation lacks explicit boundary markers or instructions to ignore embedded commands within the input data.
Capability inventory: The script has the capability to read and write to the local filesystem using the pathlib module.
Sanitization: No sanitization or validation of the input content is performed beyond standard JSON parsing.
[SAFE]: The skill uses only standard Python libraries (argparse, json, pathlib) and does not perform any network operations, external downloads, or privilege escalation. The code matches the stated purpose of a reference scaffold.

account-expansion-signal-scorer