The Agent Skills Directory

[COMMAND_EXECUTION]: The run.sh script executes the expanso command-line tool to run the data pipeline. This is the intended primary functionality of the skill.
[EXTERNAL_DOWNLOADS]: The documentation and example links in SKILL.md point to expanso.io domains. These are vendor-owned resources and are considered safe for information purposes.
[DATA_EXFILTRATION]: The pipeline.yaml configuration is designed to transmit data to external Kafka brokers and AWS S3 buckets. These destinations are configured via environment variables (KAFKA_BROKERS, S3_BUCKET), which is standard practice for infrastructure tools.
[INDIRECT_PROMPT_INJECTION]: The skill exposes a network service that ingests external data, which could potentially contain malicious instructions if processed by an agent downstream.
Ingestion points: The http_server input in pipeline.yaml listens on 0.0.0.0:8080/events to receive data.
Boundary markers: Absent; there are no specific markers or instructions to the agent to disregard embedded content in the ingested events.
Capability inventory: The skill has the capability to write to the local file system (/tmp/events.jsonl) and communicate with network-based message brokers and storage services.
Sanitization: Absent; the pipeline configuration does not define any validation or sanitization logic for the incoming HTTP request bodies.

expanso-fan-out-s3