Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent read-only Typeform helper, but users should remember it can print survey responses and account details into the agent transcript.
Install only if you are comfortable giving the skill a Typeform personal access token with the listed read scopes. Use it only on forms you are authorized to inspect, prefer narrow limits and date filters for responses, and avoid requesting raw responses when the survey may contain personal, customer, employee, health, financial, or other confidential information.
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
Forms: Read, Responses: Read, Insights: ReadTYPEFORM_TOKEN=tfp_your_token_here
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
## Setup
1. Go to https://admin.typeform.com/account#/section/tokens
2. Generate a personal access token with scopes: `Forms: Read`, `Responses: Read`, `Insights: Read`
3. Set the environment variable:
```
TYPEFORM_TOKEN=tfp_your_token_here
The skill requires environment access to a bearer token and performs direct network calls to the Typeform API, but it does not declare an explicit tool scope such as permissions or allowed-tools. This weakens least-privilege controls and makes the skill's capabilities less transparent to users and policy enforcement, increasing the chance of unintended data access or transmission.
The skill reads form responses, which commonly contain personal or sensitive respondent data, yet the description and usage guidance do not warn users about handling that data. Without a clear warning, users may invoke the skill on sensitive surveys without understanding the privacy implications of retrieving and displaying respondent information from the API.
The responses command prints raw form answers directly to stdout, which can expose sensitive personal or confidential survey data to logs, transcripts, or downstream tooling without any warning or minimization. In this skill context, response retrieval is expected, but unrestricted display of all answer content materially increases privacy and data-handling risk.
The skill exposes a me command that retrieves account-profile information, which is outside the stated scope of reading forms, responses, and performance stats. While not inherently malicious, this broadens accessible data and violates least-privilege expectations for the skill.
Account-info lookup is not justified by the declared skill purpose and may reveal metadata about the user's Typeform account that callers would not expect this integration to access. The issue is a scope mismatch rather than an exploit primitive, but it still increases unnecessary data exposure.