The Agent Skills Directory

[PROMPT_INJECTION]: The skill presents a vulnerability surface for indirect prompt injection.
Ingestion points: Untrusted data enters the agent context via the WebFetch tool from external websites as described in the skill capabilities.
Boundary markers: The instructions fail to provide delimiters or explicit "ignore-instructions" warnings to isolate the processed external data from the agent's core instruction set.
Capability inventory: The skill configuration in SKILL.md grants access to the Bash, Write, Read, and WebFetch tools.
Sanitization: No sanitization, filtering, or validation logic is defined for the content retrieved from external sources before it is processed into JSON tables.

nimble