The Agent Skills Directory

[SAFE]: The skill provides standard boilerplate code and architectural patterns for browser extension development using Manifest V3, focusing on educational templates.
[PROMPT_INJECTION]: No evidence of prompt injection, jailbreak attempts, or instructions to bypass safety guidelines was found in the instructions or metadata.
[DATA_EXFILTRATION]: The skill demonstrates the use of standard chrome.storage APIs for local data persistence and does not contain any patterns for network exfiltration or sensitive data harvesting.
[COMMAND_EXECUTION]: No shell commands, subprocess calls, or system-level execution patterns were identified in the provided templates.
[EXTERNAL_DOWNLOADS]: The skill does not reference or download any external scripts, binaries, or unverifiable dependencies from remote sources.
[OBFUSCATION]: All content is provided in clear text with no evidence of Base64 encoding, zero-width characters, homoglyphs, or other obfuscation techniques.
[PRIVILEGE_ESCALATION]: The skill explicitly promotes security best practices by including an anti-pattern warning against requesting excessive permissions, advocating for the principle of least privilege.
[INDIRECT_PROMPT_INJECTION]: While the skill provides templates for content scripts that ingest data from web pages (document.querySelector), it does not pair this with dangerous capabilities like network exfiltration or dynamic execution, presenting no significant risk.

browser-extension-builder