Security audit
Security checks for vulnerabilities and agentic risk
This skill manages social-space photo albums, but it relies on full-access session cookies and non-official APIs with incomplete warnings about how those credentials are used.
Review this skill carefully before installing. Only use it with an account you are comfortable automating, keep cookies.json private with restrictive file permissions, avoid backing up or sharing cookie files, and refresh or revoke the session if the file may have been exposed. Confirm each upload, album creation, or full-album download before letting an agent execute it.
The file presents the skill as pure Markdown/natural-language driven while elsewhere declaring exec capability and providing shell commands to run. This mismatch can mislead users and reviewers about the actual execution surface, causing them to underestimate the risk of command execution and local side effects.
The security section claims credentials stay local and are not uploaded to any server, but the skill's core function is to use session cookies to authenticate to external non-official APIs. That statement is materially misleading because the cookies or derived authenticated requests must be transmitted off-host, which can cause users to disclose high-privilege session tokens under false assumptions.
The documentation describes upload, download, and album creation workflows without clearly warning that these actions will modify remote account data and write copies of user content to the local filesystem. In an agent context, omission of these side effects increases the chance of unintended data changes, privacy exposure, or accidental bulk copying.
The skill instructs users to obtain and store reusable session cookies, including manual extraction from browser developer tools, without a sufficiently prominent warning that these tokens can grant full account access if copied or leaked. In this context, the cookies function as high-value bearer credentials, so encouraging extraction and reuse materially raises account-compromise risk.
No suspicious patterns detected.
