The Agent Skills Directory

[PROMPT_INJECTION]: No attempts to override system instructions, bypass safety filters, or extract system prompts were found. The instructions are strictly limited to defining the AI's behavior for creating interview guides.
[DATA_EXPOSURE]: No sensitive file paths, hardcoded credentials, or network exfiltration patterns were identified. The skill does not access the local file system (beyond its own referenced markdown files) or environment variables.
[REMOTE_CODE_EXECUTION]: The skill consists entirely of markdown documentation and templates. It does not attempt to download, install, or execute any external scripts or packages.
[OBFUSCATION]: All content is provided in plain text. No Base64 encoding, zero-width characters, homoglyphs, or other techniques to hide malicious content were detected.
[DYNAMIC_CONTEXT_INJECTION]: The skill does not use the !command syntax or any other mechanism to execute shell commands at load time.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests user-provided research topics and profiles, it possesses no exploitable capabilities (such as network access, file writing, or command execution) that could be leveraged if malicious instructions were embedded in the input.

discovery-interview-guide