The Agent Skills Directory

[DATA_EXFILTRATION]: The skill contains methodologies and an automated script (scripts/ssrf_agent.py) designed to access sensitive data. It specifically targets cloud metadata endpoints to extract IAM security credentials and local system files such as /etc/passwd and /proc/self/environ using the file:// protocol handler.
[COMMAND_EXECUTION]: The Python script (scripts/ssrf_agent.py) executes automated network operations, including port scanning for services like Docker and Kubernetes, and making HTTP requests to internal or external targets provided by the user.
[CREDENTIALS_UNSAFE]: The instructions and scripts focus on the extraction of sensitive authentication tokens. This includes capturing AccessKeyId, SecretAccessKey, and Token from AWS IMDS endpoints, as well as OAuth tokens from GCP and Azure managed identities.
[INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by fetching and processing data from external, untrusted URLs without sanitization.
Ingestion points: The test_ssrf_payload function in scripts/ssrf_agent.py retrieves content from arbitrary URLs using requests.get.
Boundary markers: None identified; the agent is instructed to process and report the raw response body.
Capability inventory: Network operations, internal network probing, and reading sensitive local files.
Sanitization: No evidence of filtering or validation of the retrieved content before it is presented to the agent.
[EXTERNAL_DOWNLOADS]: The skill documentation recommends the installation of the requests Python package, which is a standard library for handling HTTP requests.

ssrf-hunter