The Agent Skills Directory

[PROMPT_INJECTION]: The skill ingests and processes source code from the local environment, which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Content is retrieved from the filesystem using tools like get_code_snippet, search_code, and search_graph.
Boundary markers: The instructions lack explicit delimiters or guidance for the agent to treat retrieved code as data only and to ignore any embedded instructions.
Capability inventory: The agent is empowered to use information from these tools to plan complex architectural changes and execute file modifications.
Sanitization: There is no evidence of filtering or validation of the code snippets retrieved from the graph to prevent execution of embedded prompts.

code-graph