The Agent Skills Directory

[SAFE]: No security issues were detected. The skill focuses on enforcing a citation format and verifying facts through standard tools such as read, grep, and bash. The instructions are transparent and aligned with the stated purpose of improving response quality through evidence chaining.
[INDIRECT_PROMPT_INJECTION]: The skill inherently possesses an attack surface for indirect prompt injection because its primary function involves fetching and processing data from external sources (files, shell outputs, and URLs) to verify claims.
Ingestion points: The skill instructs the agent to ingest untrusted data via read, grep, and fetching external documents (doc: <URL>).
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are defined for the external data being processed.
Capability inventory: The agent is authorized to use bash for system checks, read for file access, and network operations to fetch research documents.
Sanitization: There are no explicit sanitization or filtering instructions for the content retrieved from external sources. However, given the skill's primary purpose is auditing and fact-checking, this risk is inherent to its operation and managed by the agent's underlying safety guardrails.

province-chain