The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from remote target responses during security testing, which is an inherent part of its functionality.
Ingestion points: The skill captures and analyzes HTTP response bodies, status codes, and headers via curl and ffuf as described in Phases 1 through 6 of the methodology.
Boundary markers: No explicit instructions or delimiters are provided to the agent to ignore potentially malicious instructions embedded in the target's HTTP responses.
Capability inventory: The skill utilizes network access (via curl, ffuf, hydra, and nuclei) and shell command execution to perform its auditing tasks.
Sanitization: The methodology does not specify sanitization or filtering of the content received from the target servers before it is processed by the agent.
[DYNAMIC_EXECUTION]: The skill generates and executes local shell scripts and Python code at runtime to perform specific auditing calculations and automated probing.
Evidence: The skill utilizes bash for loops for request bursting and python3 -c snippets in Phase 5 to calculate Shannon entropy and analyze token structure at runtime.

hunt-brute-force