The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses natural language instructions to guide the agent in offer construction. It does not contain any markers or patterns associated with prompt injection, such as instructions to ignore previous rules, bypass safety filters, or reveal system prompts.
[DATA_EXFILTRATION]: No network operations, sensitive file path access, or credential harvesting patterns were detected. The skill's recommendation to read local context files like .agents/product-marketing.md is a standard practice for maintaining project context and does not involve data exfiltration.
[REMOTE_CODE_EXECUTION]: There are no commands that download or execute external scripts, nor does the skill use tools like curl, wget, or shell pipes to run remote content.
[COMMAND_EXECUTION]: The skill is entirely composed of markdown documentation and instructional text. It does not contain any shell commands, subprocess calls, or scripts.
[OBFUSCATION]: A thorough scan of the files revealed no Base64-encoded strings, zero-width characters, homoglyphs, or hidden acrostics designed to conceal malicious intent or URLs.
[CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or secrets were found. The skill does not instruct the agent to handle or store credentials in an unsafe manner.
[EXTERNAL_DOWNLOADS]: The skill does not attempt to download external packages or assets from the internet. All references are to internal documentation within the references/ directory.
[SAFE]: The content is purely educational, providing frameworks for offer design, guarantee structures, and value positioning.

offers