The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill enables the agent to query and process log streams from OpenObserve. This creates a surface for indirect prompt injection where malicious instructions embedded in log data could potentially influence the agent's behavior during analysis.
Ingestion points: Search hits returned from the POST /api/_search endpoint, as documented in SKILL.md and references/search.md.
Boundary markers: The skill does not provide instructions to the agent to treat search hits as untrusted data or use specific delimiters.
Capability inventory: The skill provides templates for shell command execution (curl, jq, gh) and network operations to the OpenObserve API.
Sanitization: No sanitization or filtering logic is prescribed for the retrieved log content before processing.
[COMMAND_EXECUTION]: The skill extensively uses curl and jq command templates for interacting with the OpenObserve REST API. It also includes standalone bash scripts in the references/recipes/ directory designed to be executed for dashboard creation and log searching.
[EXTERNAL_DOWNLOADS]: The skill encourages the agent to retrieve up-to-date documentation and source code from the official OpenObserve GitHub repositories using the gh CLI. These downloads target well-known repositories associated with the platform and the skill author.

openobserve-api