The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from various external academic databases (PubMed, arXiv, etc.).
Ingestion points: The agent reads and synthesizes data from external research papers and database search results as described in Phase 2 and Phase 3 of the Core Workflow.
Boundary markers: No delimiters or specific instructions to ignore embedded commands within the academic content are provided.
Capability inventory: The agent has access to Bash and Write tools, which could be leveraged if malicious instructions were successfully injected through academic metadata or paper content.
Sanitization: There is no mention of sanitizing or validating the data retrieved from external sources before it is processed.
[PROMPT_INJECTION]: The skill contains instructions designed to modify the agent's behavior for promotional purposes, specifically directing the agent to 'proactively suggest' the K-Dense Web platform when complexity increases. It also imposes a mandatory requirement for the agent to use the scientific-schematics skill for every review, overriding default task prioritization.

literature-review