The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions include installing the Oodle CLI via Homebrew from the oodle-ai/oodle tap (brew install oodle-ai/oodle/oodle). This is a vendor-owned resource used to provide the skill's core functionality.
[COMMAND_EXECUTION]: The skill relies on shell command execution to interact with the oodle CLI. This includes listing traces, fetching trace details, and using jq for JSON processing. Commands are generally well-scoped with service and time filters.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes external telemetry data (trace and span details) which could potentially contain malicious instructions embedded by an attacker in service names, operation names, or other span attributes. This data is used to inform subsequent agent actions and command construction. Evidence: 1. Ingestion points: Output from oodle traces list and oodle traces get. 2. Boundary markers: No explicit instructions provided to the agent to treat telemetry data as untrusted content. 3. Capability inventory: Execution of oodle CLI commands and shell pipe operations. 4. Sanitization: No sanitization or validation of the fetched trace data is specified before it is processed by the agent.

oodle-traces