The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is instructed to ingest data from the target repository by reading project-specific guidance files like AGENTS.md, SKILL.md, and README.md. This creates an attack surface where a malicious repository could influence the agent's behavior via embedded instructions, though the skill directs the agent to verify all behavior against backend sources of truth.\n- [COMMAND_EXECUTION]: The skill directs the agent to discover and execute verification and testing commands (such as npm test, make verify, or gradlew verify) defined within the repository's configuration. This involves executing arbitrary shell commands based on the content of the workspace, which is a functional requirement for SDK development that requires user trust in the target repository.

putio-sdk-dev