The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the retrieval and processing of content from external sources, which presents a potential surface for indirect instructions to enter the agent's context.
Ingestion points: The skill instructions specify fetching content from official Esri documentation domains (developers.arcgis.com, pro.arcgis.com, enterprise.arcgis.com) and searching the Esri/arcgis-python-api GitHub repository (found in SKILL.md).
Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions within the fetched data.
Capability inventory: The skill suggests using the python-notebook skill for introspection and potentially escalating complex tasks to a research skill (found in SKILL.md).
Sanitization: The instructions do not define specific sanitization or validation steps for the retrieved external content.
[DYNAMIC_EXECUTION]: The skill describes using Python code execution for the purpose of API verification and package introspection.
Evidence: The instructions suggest running Python commands such as help(Group.notify) or inspect.signature(...) within a live environment (e.g., using a python-notebook skill) to confirm API signatures when documentation pages fail to render (found in SKILL.md).

arcgis-docs-lookup