The Agent Skills Directory

[SAFE]: The skill provides standard research methodology and debugging workflows. No evidence of credential exposure, data exfiltration, or malicious commands was found.
[INDIRECT_PROMPT_INJECTION]: The skill's architecture presents a vulnerability surface for indirect prompt injection.
Ingestion points: The agent gathers failure cases, results, and qualitative observations from user-provided logs and the research environment.
Boundary markers: The instructions lack explicit boundary delimiters to prevent the agent from following instructions embedded within external experimental data.
Capability inventory: The agent is equipped with write_file, edit_file, read_file, and execute tools.
Sanitization: The skill does not implement or recommend sanitization or validation of data before it is processed by the agent.

experiment-craft