The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches OpenTelemetry semantic convention schemas from the official OpenTelemetry GitHub repositories (api.github.com and raw.githubusercontent.com). These downloads are used to perform schema drift analysis and validate that the service telemetry conforms to industry standards.
[COMMAND_EXECUTION]: The skill instructs the agent to execute various observe CLI subcommands (e.g., observe query, observe dataset, observe auth) to interact with the Observe platform. It also utilizes standard shell utilities such as curl, jq, sort, and tail to retrieve and parse schema information from remote sources.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests and processes telemetry data and external schema files.
Ingestion points: The skill reads telemetry data output from observe query and YAML-formatted schema files fetched via curl in references/instrumentation-audit.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded content are used when processing telemetry fields or schema content.
Capability inventory: The skill utilizes shell command execution (the observe CLI, curl, and jq) to perform its primary functions.
Sanitization: The skill relies on jq for parsing JSON output and standard shell processing for derived data points, though no specific sanitization for natural language instructions within the data is implemented.

opentelemetry-validation