The Agent Skills Directory

[PROMPT_INJECTION]: No attempts to override agent behavior, bypass safety guidelines, or extract system prompts were detected in the instructions or evaluation prompts.
[DATA_EXFILTRATION]: The skill does not access sensitive file paths or perform network operations. It mentions reading standard project context files (e.g., .agents/product-marketing-context.md) to provide tailored advice, which is a common and safe practice for context-aware agents.
[REMOTE_CODE_EXECUTION]: There are no scripts, package dependencies, or remote download commands included in the skill.
[OBFUSCATION]: All content is in plain text. No Base64, zero-width characters, or other encoding techniques were found.
[COMMAND_EXECUTION]: No shell commands or dynamic context injections (!command syntax) are present in the skill files.
[INDIRECT_PROMPT_INJECTION]: While the skill reads context from external files, it lacks the exploitable capabilities (like file-writing or network sending) that would typically characterize a significant indirect injection risk.

free-tool-strategy