The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external worker logs and command-line outputs to perform layer-by-layer diagnostics.
Ingestion points: Worker logs (referenced in SKILL.md and references/triage/worker-health.md) and copy-pasted CLI error strings are used as primary data sources for troubleshooting.
Boundary markers: Instructions do not provide explicit delimiters or "ignore instructions" wrappers for the ingested log content, relying on standard gRPC error prefixes for classification.
Capability inventory: The skill utilizes administrative CLI tools (temporal, tcld) with capabilities for destructive operations such as namespace deletion and workflow termination (SKILL.md, references/ops/cloud-namespace-admin.md).
Sanitization: No explicit sanitization or filtering logic is prescribed for the external log data before it enters the agent context.
[EXTERNAL_DOWNLOADS]: The skill refers to and provides instructions for downloading vendor-provided tools and dependencies from trusted infrastructure.
Fetches the Temporal and Temporal Cloud CLI tools via the Homebrew package manager (SKILL.md).
References official Temporal repositories on GitHub for Cloud API definitions and SDK samples (references/ops/cloud-ops-api.md).
Downloads AWS CloudFormation templates for audit log integration from a vendor-managed S3 bucket (references/ops/cloud-audit-logs.md).

temporal-ops