The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches version information and updated instruction files (SKILL.md) from skills.opper.ai and the opper-ai GitHub organization. These resources are hosted on the vendor's official infrastructure.
[COMMAND_EXECUTION]: The skill uses shell commands to inspect the local filesystem for environment variables (OPPER_API_KEY) and existing project dependencies in files such as package.json, pyproject.toml, and requirements.txt to determine the best integration path.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes local application source code to identify existing third-party LLM integrations. This analysis creates an attack surface where instructions embedded in the project files could potentially influence the agent's routing decisions.
Ingestion points: Phase 1 grep operations on local source files.
Boundary markers: None specified for the analysis of file content.
Capability inventory: Shell command execution, network requests via curl, and package management via npx.
Sanitization: No explicit sanitization of the content found during file inspection is mentioned.
[DYNAMIC_EXECUTION]: The skill implements a self-update mechanism that instructs the agent to fetch and follow remote instructions from the vendor's website if the local copy is detected to be out of date.

opper