The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to autonomously research, read, and synthesize information from external sources. This creates a surface for indirect prompt injection where malicious instructions embedded in web pages or documents could influence the agent's behavior.
Ingestion points: External web content and search results retrieved and processed by the scripts/research.py script.
Boundary markers: The skill does not define specific delimiters or "ignore instructions" markers for the data it processes.
Capability inventory: Shell command execution for running Python scripts and network access via the httpx library.
Sanitization: There are no specific sanitization or validation protocols described for the external content being ingested.
[COMMAND_EXECUTION]: The skill instructions direct the agent to execute shell commands using python3 scripts/research.py with various arguments. This functionality relies on the integrity of the local script file within the skill package.
[EXTERNAL_DOWNLOADS]: The skill requires installing dependencies from a requirements.txt file. While the documentation mentions httpx, the full content of the requirements file is not provided, posing a minor risk of installing unverified third-party packages.

deep-research