The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute a wrapper script (burp-search.sh) and interact with the Java-based Burp Suite Professional executable for project parsing.
[EXTERNAL_DOWNLOADS]: The documentation instructs the user to manually download and install a JAR file from an external third-party repository (github.com/BuffaloWill/burpsuite-project-file-parser) to enable CLI functionality.
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks as it processes untrusted HTTP traffic data contained within Burp project files.
Ingestion points: Data enters the agent context through the parsing of .burp files via the burp-search.sh script referenced in SKILL.md.
Boundary markers: The skill explicitly defines safety boundaries, such as mandatory output truncation using head -c 50000 and jq to limit response body fields to 1000 characters.
Capability inventory: The skill has the capability to execute shell commands and file read operations using the Bash and Read tools.
Sanitization: It relies on user-invoked jq filters and manual truncation commands to manage and sanitize the output from external files before it is processed by the agent.

burpsuite-project-parser