The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill involves scanning and analyzing data from external, untrusted GraphQL endpoints and JavaScript bundles, which could potentially contain malicious instructions aimed at the agent. * Ingestion points: Target GraphQL endpoint responses, schema definitions, and JS bundle contents. * Boundary markers: Not present. The skill does not instruct the agent to use specific delimiters or ignore embedded instructions when analyzing target data. * Capability inventory: The skill utilizes network tools (curl), pattern matching (grep), and security scanners (inql, graphql-cop, clairvoyance.py) to process data. * Sanitization: The instructions do not specify sanitization or validation steps for the content retrieved from target environments.
[COMMAND_EXECUTION]: The skill requires the agent to execute various shell commands and CLI tools to perform its intended security research functions, including curl for HTTP requests, grep for searching JS bundles, and specialized Python scripts or binaries for schema auditing.

hunt-graphql