The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves reading external files (registry.json and flywheel-connections.md) and incorporating their content into its reasoning process, which creates a surface for indirect instructions.
Ingestion points: Untrusted data enters the agent context via the registry.json file in the root and the shared/references/flywheel-connections.md file.
Boundary markers: The workflow does not define any delimiters or system-level instructions to ignore potential commands embedded within the registry entries.
Capability inventory: The skill does not use any tools or scripts that perform network requests, file writes, or shell command execution.
Sanitization: There is no evidence of input validation, filtering, or escaping for the data retrieved from the registry files.

skill-finder