The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze log.txt, a file containing captured network traffic. Since this data originates from external sources, it represents a potential surface where an attacker could embed malicious instructions within HTTP headers or body content to influence the AI agent's analysis.
Ingestion points: Reads log.txt (mitmproxy dump) as the primary data source.
Boundary markers: No specific delimiters or "ignore instructions" warnings are present to isolate the log content from the analysis logic.
Capability inventory: The skill instructs the agent to use grep for searching patterns and suggests curl for testing identified vulnerabilities (SKILL.md).
Sanitization: The skill does not implement sanitization or validation of the traffic data before analysis.
[COMMAND_EXECUTION]: The skill provides several bash code snippets involving grep and curl. These are presented as manual diagnostic tools and testing steps for the user to verify logic flaws like price manipulation or race conditions in a controlled environment.

mitm-find-bizlogic