The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests research data from the OpenRouter API and presents it directly to the agent's context, which could allow for indirect prompt injection if external sources contain malicious instructions.\n
Ingestion points: External data is ingested in the lookup method within both research_lookup.py and scripts/research_lookup.py via API responses from OpenRouter.\n
Boundary markers: Absent. The skill does not wrap the retrieved external content in delimiters or provide instructions to the agent to disregard commands within the research data.\n
Capability inventory: The skill allows the agent to use Bash, Read, Write, and Edit tools as defined in SKILL.md.\n
Sanitization: Absent. There is no evidence of escaping, filtering, or validation performed on the text returned from the external API before it is presented to the agent.

research-lookup