The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides various shell commands using curl, aws-cli, and docker to perform infrastructure reconnaissance and validation.
Evidence includes S3 bucket listing commands, Firebase database testing, and LocalStack environment setup.
[EXTERNAL_DOWNLOADS]: The skill includes instructions to pull the localstack/localstack:3.0 container image from Docker Hub.
This is a well-known service for local cloud emulation and is used here for safe verification of findings.
[INDIRECT_PROMPT_INJECTION]: The skill represents an indirect prompt injection attack surface as it is designed to ingest and process data from external, untrusted sources.
Ingestion points: S3 buckets, Firebase databases, and HTTP responses from scanned endpoints (SKILL.md).
Boundary markers: None identified in the provided instructions.
Capability inventory: The skill uses curl for network reads/writes, aws-cli for cloud operations, and docker for container management.
Sanitization: There is no explicit sanitization or filtering logic for the content read from external targets before the agent processes it.

hunt-cloud-misconfig