The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script (last30days.py) using python3 and passes user-provided $ARGUMENTS directly into the shell command.
[PROMPT_INJECTION]: The _scores.json file contains a pre-defined security verdict of SAFE. This is a self-referential claim designed to influence security evaluations and is disregarded in favor of independent analysis.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting untrusted user input through the TOPIC and $ARGUMENTS variables.
Ingestion points: User input parsed in SKILL.md (Step 1 and Step 3).
Boundary markers: No delimiters or instructions are provided to the agent to treat interpolated user input as untrusted or to ignore embedded instructions.
Capability inventory: The skill possesses the ability to execute shell commands via python3 and write files to the user's home directory (~/Documents/Last30Days).
Sanitization: There is no evidence of input validation, escaping, or sanitization before the data is passed to the shell or the search engine.

last30days-skill