The Agent Skills Directory

[NO_CODE]: The skill is composed exclusively of Markdown instructions. It does not include any scripts (Python, JavaScript, Bash), binary executables, or environment configuration files.
[SAFE]: No malicious patterns such as prompt injection, credential harvesting, or obfuscation were detected. The instructions focus on a systematic 100-point rubric for evaluating landing page messaging and conversion strategies.
[PROMPT_INJECTION]: While the skill involves processing external content (landing pages and screenshots) which is a surface for Indirect Prompt Injection, the skill does not provide the agent with any dangerous tools or permissions (such as file-system write access or network exfiltration tools) that could be abused by an external attacker. The risk is limited to the agent potentially receiving biased instructions from a target website during the analysis phase.

competitor-teardown