← Back to Skills Marketplace
jonasboury

Unformal Api

by Jonas Boury · GitHub ↗ · v1.4.0 · MIT-0
cross-platform ⚠ suspicious
206
Downloads
0
Stars
1
Active Installs
3
Versions
Install in OpenClaw
/install unformal-api
Description
Create conversational Pulses that replace forms, surveys, intake emails, feedback requests, NPS checks, user interviews, lead qualification, application form...
Usage Guidance
This skill appears to do what it says: it talks to unformal.ai and uses the Unformal CLI/API to create Pulses. Before installing or running it, consider: (1) verify you trust https://unformal.ai and the npm package @unformal/cli (review package and version) because the skill suggests installing it globally; (2) the login flow issues an API key (unf_...) which the skill recommends saving to ~/.unformal/config — treat that like any API secret (do not reuse high-privilege keys elsewhere); (3) the SKILL.md can overwrite the agent's local skill file (~/.claude/skills/unformal/SKILL.md) when updating — back up any local changes you care about; (4) network calls go to unformal.ai only; if you need stricter controls, avoid global installs and create a dedicated, limited Unformal account/API key for the integration. Overall the skill is internally consistent, but you should validate the external CLI/package before running installs or saving credentials.
Capability Analysis
Type: OpenClaw Skill Name: unformal-api Version: 1.4.0 The skill contains instructions in SKILL.md for the AI agent to perform self-updates by executing 'npm i -g @unformal/cli@latest' and overwriting its own instruction file via 'curl -fsS https://unformal.ai/SKILL.md'. This behavior creates a significant risk of Remote Code Execution (RCE) and instruction injection, as the remote server can change the agent's logic or install arbitrary software at any time. While these actions are documented as a way to maintain API compatibility, the ability to modify the execution environment and instructions from a remote source is a high-risk pattern that bypasses static analysis.
Capability Tags
requires-sensitive-credentials
Capability Assessment
Purpose & Capability
The name/description describe creating conversational Pulses and the SKILL.md only instructs use of the Unformal API and CLI to create pulses, fetch responses, and manage config. Required resources (CLI via npm, Unformal API key) align with the purpose; there are no unrelated credentials, binaries, or surprising system access requests.
Instruction Scope
Runtime instructions are limited to contacting https://unformal.ai endpoints, using the @unformal/cli (optional), and saving/reading the user's Unformal API key (e.g., ~/.unformal/config). The file also suggests updating the SKILL.md by downloading from unformal.ai into the agent's skills directory — this is self-contained and consistent with keeping the skill current. It does not instruct reading unrelated system files or exfiltrating data to unknown endpoints.
Install Mechanism
There is no install spec in the registry (instruction-only). The skill suggests installing the CLI with npm i -g @unformal/cli if needed — this is a common and expected mechanism but does mean external code from the npm registry will be installed; users may want to review the npm package before installing.
Credentials
The skill does not declare required env vars in the registry, but its documented flow uses an Unformal API key (unf_...) saved to ~/.unformal/config or passed in Authorization headers. Requesting and storing an API key for the service is proportional and expected; there are no unrelated secret requests.
Persistence & Privilege
always is false and the skill is user-invocable (normal). The SKILL.md suggests writing to ~/.claude/skills/unformal/SKILL.md to update the skill file and saving the API key to ~/.unformal/config — both are limited to the skill's own files/config and are explainable by the update/login flows. Users should be aware that the skill recommends global npm installation (npm i -g) which affects the system environment.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install unformal-api
  3. After installation, invoke the skill by name or use /unformal-api
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.4.0
Freshness instruction — skill now tells agents to check /api/v1/version + refresh before creating a Pulse.
v1.3.0
Expand use-case triggers (feedback, NPS, surveys, interviews, lead qualification, intake, HR). Add When-to-use section. Add login flow (POST /api/v1/login + /login/verify) for existing users.
v1.0.0
Initial release: create Pulses, manage conversations, get structured insights via API. Agent signup, email verification, webhooks.
Metadata
Slug unformal-api
Version 1.4.0
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 3
Frequently Asked Questions

What is Unformal Api?

Create conversational Pulses that replace forms, surveys, intake emails, feedback requests, NPS checks, user interviews, lead qualification, application form... It is an AI Agent Skill for Claude Code / OpenClaw, with 206 downloads so far.

How do I install Unformal Api?

Run "/install unformal-api" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Unformal Api free?

Yes, Unformal Api is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Unformal Api support?

Unformal Api is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Unformal Api?

It is built and maintained by Jonas Boury (@jonasboury); the current version is v1.4.0.

💬 Comments