The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches instrumentation source code from the official OpenTelemetry JavaScript Contrib repository (open-telemetry/opentelemetry-js-contrib) using the GitHub API.
[COMMAND_EXECUTION]: Utilizes system commands such as gh, grep, and pnpm to query the environment, fetch remote assets, and execute local build and test suites to verify integration.
[INDIRECT_PROMPT_INJECTION]: The skill processes external source code from a third-party repository, which represents a potential attack surface for indirect prompt injection.
Ingestion points: Source code fetched via gh api from the open-telemetry/opentelemetry-js-contrib repository (file: SKILL.md).
Boundary markers: The skill does not explicitly define delimiters for the fetched code content, but it requires the agent to present findings and diffs to the user.
Capability inventory: The skill has capabilities for file system modifications, network access via the gh tool, and execution of build/test scripts via pnpm (file: SKILL.md).
Sanitization: Employs mandatory manual review checkpoints where the user must approve the plan, verify diffs, and confirm PR creation before the agent proceeds (file: SKILL.md).

vendor-otel