The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to perform environment discovery and dependency identification using commands like ls -a.
[PROMPT_INJECTION]: The skill ingests and analyzes local source code and configuration files, creating a surface for indirect prompt injection if the audited codebase contains malicious instructions. \n
Ingestion points: Local project files, including source code, READMEs, and configuration manifests (e.g., package.json, pyproject.toml). \n
Boundary markers: No delimiters or protective instructions are used to isolate ingested content from the assessment logic. \n
Capability inventory: Access to Bash for environment detection, Read for file content access, and WebFetch for potential external resource validation. \n
Sanitization: No content validation or sanitization steps are documented before processing the gathered data.
[DATA_EXFILTRATION]: Architectural discovery data, including detailed endpoint maps and authentication logic specifications, is consolidated into an HTML report via the /atlas-report command.

spine-recon