The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses natural instructional language to define a professional persona (senior product analyst) and a structured workflow. It does not contain bypass markers, jailbreak attempts, or instructions to ignore safety protocols.
[DATA_EXFILTRATION]: No sensitive file paths, hardcoded credentials, or unauthorized network operations were identified. The skill's primary output is writing documentation to a local directory (specs/).
[REMOTE_CODE_EXECUTION]: There are no patterns indicating the download or execution of remote scripts or external packages.
[COMMAND_EXECUTION]: The skill does not invoke shell commands, subprocesses, or administrative tools. It primarily interacts via the AskUserQuestions tool and file system writes for documentation.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input via the AskUserQuestions tool to generate specifications. While it lacks explicit delimiters to wrap user input, the output is limited to Markdown documentation, which does not constitute an exploitable capability for runtime execution.
[DYNAMIC_EXECUTION]: No code generation, runtime compilation, or unsafe deserialization patterns were found. The skill operates on static templates and structured user responses.

feature-forge