The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided source code for analysis and documentation, which serves as a surface for indirect prompt injection. Malicious instructions could be embedded in the codebase to influence the agent's behavior during the audit preparation process.
Ingestion points: Source code files are read from the repository during static analysis and documentation generation steps.
Boundary markers: The instructions do not specify any markers or techniques to isolate the codebase content from the agent's control logic.
Capability inventory: The skill instructs the agent to execute shell commands such as slither, dylint, golangci-lint, and forge.
Sanitization: No explicit sanitization or validation of the codebase content is described before processing.
[COMMAND_EXECUTION]: The skill involves the execution of various command-line interface tools to perform security checks and verify builds. These tools (Slither, Dylint, GolangCI-Lint, Forge) are standard industry utilities for the respective programming languages.
Evidence: Execution of static analysis tools and build commands defined in the 'The Preparation Process' and 'How I Work' sections of SKILL.md.

audit-prep-assistant