Security audit
Security checks for vulnerabilities and agentic risk
This skill does what it says: it uploads a user-provided encrypted PDF and password to PDFAPIHub to return an unlocked PDF, with the main risk clearly disclosed.
Only use this for PDFs you are allowed to unlock, and avoid highly sensitive documents unless you accept sending both the document and its password to PDFAPIHub. Use secure handling for the API key and password instead of copying real secrets into examples or saved files.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST https://pdfapihub.com/api/v1/unlockPdf \
-H "CLIENT-API-KEY: your_api_key" \
-H "Content-Type: application/json" \
-d '{
The example hard-codes values that look like real secrets: an API key field populated with "your_api_key" and a concrete password value "secret123". Even if intended as illustrative placeholders, embedding credential- and password-shaped literals in examples can normalize unsafe secret handling, lead users to copy sensitive data into manifests, and increase accidental exposure in logs, version control, or support artifacts.
The skill sends highly sensitive material to a third-party service: the encrypted PDF itself and the password needed to decrypt it. Although the description mentions that data is sent to pdfapihub.com, it does not present an explicit privacy/security warning, retention statement, or user-consent mechanism at the point of use, so users may unknowingly transmit confidential documents and credentials to an external processor.
This JSON manifest appears to define a skill action but provides no indication of when it should be invoked, what user intents it is meant to handle, or any exclusion conditions. For manifest files, missing trigger specificity can cause unintended invocation because the action is described only by a generic API call structure.