100Critical
This instructional skill describes using `curl` and `jq
Behavior Analysis
Mismatch detected — The skill's stated purpose implies it 'enables interaction' and 'uses `curl` and `jq`' for direct REST calls. However, the content is purely instructional, providing examples of `curl` commands. The skill itself does not execute any commands or interact with the Notion API; it only describes how a user or another system *could* do so.
Claims to do
Notion API Skill: This skill enables interaction with Notion workspaces through the Notion REST API. Use `curl` and `jq` for direct REST calls, or write ad-hoc scripts as appropriate for the task.
Actually does
This skill provides detailed instructions and `curl` examples for interacting with the Notion REST API. It outlines how to use `curl` and `jq` to perform operations like searching, creating, retrieving, updating, and deleting Notion pages, blocks, databases, data sources, users, and comments. It requires a `NOTION_API_TOKEN` (from environment or user input) and contacts `https://api.notion.com/v1/*` endpoints.
Skill Info
Nameintellectronica/agent-skills/notion-api
Registrygithub
Version9b0e00ad1b94
PURLpkg:github/intellectronica/agent-skills@9b0e00ad1b94?skill=notion-api
Stars257
Installs22,500
Source
SHA-256cad2c5f98f06...
SHA-1ba53338414bd...
MD5de0e1e1e21b8...
TLSHf462b676b461...
ssdeep192:cwuXOvYi...
Size14,675 bytes
Install
Assessments (4)
AI Agent Traps ↗Command Execution via curl and jqcriticalcommand execution
The skill explicitly instructs the agent to use `curl` and `jq` for direct REST calls, implying shell command execution capability. This allows an attacker to potentially execute arbitrary commands if prompt input is not properly sanitized.
100% confidenceline 12
Use `curl` and `jq` for direct REST calls, or write ad-hoc scripts as appropriate for the task.
Potential for Credential Theft via Command Manipulationhighcredential theft
The Notion API token (`$NOTION_API_TOKEN`) is directly used in `curl` commands. If these commands can be manipulated via prompt injection, the token could be exposed or exfiltrated to an attacker-controlled endpoint.
90% confidenceline 29
-H "Authorization: Bearer $NOTION_API_TOKEN"
SSRF and Data Exfiltration via curlhighssrf, data exfiltration
The skill's reliance on `curl` for making HTTP requests presents a risk for Server-Side Request Forgery (SSRF) or data exfiltration if an attacker can manipulate the target URL or request body through prompt injection.
90% confidenceline 95
curl -s -X POST "https://api.notion.com/v1/search"
Description Mismatchhighdescription mismatch
The skill's stated purpose implies it 'enables interaction' and 'uses `curl` and `jq`' for direct REST calls. However, the content is purely instructional, providing examples of `curl` commands. The skill itself does not execute any commands or interact with the Notion API; it only describes how a user or another system *could* do so.
85% confidence
The skill content consists entirely of markdown documentation, explanations, and `curl` command examples. There are no executable scripts or functions within the skill that would directly run `curl` or `jq`.
Bypass of Human Confirmation for Destructive Operationsmediumoversight evasion
The skill relies on the agent to ask for user confirmation before executing destructive operations. This is a guideline that could be bypassed by a malicious prompt, leading to unauthorized data modification or deletion.
80% confidenceline 79
IMPORTANT: Before executing any operation that modifies or deletes data, ask the user for confirmation.
Credential Harvesting Mechanismmediumcredential harvesting
The skill instructs the agent to request the API key from the user if it's not available. This mechanism could be exploited by a malicious prompt to harvest credentials, especially when combined with command execution capabilities.
80% confidenceline 20
If neither is available, use AskUserQuestion (or equivalent) to request the API key from the user
Badge
Markdown
[](https://mondoo.com/ai-agent-security/skills/github/intellectronica/agent-skills/notion-api)HTML
<a href="https://mondoo.com/ai-agent-security/skills/github/intellectronica/agent-skills/notion-api"><img src="https://mondoo.com/ai-agent-security/api/badge/github/intellectronica/agent-skills/notion-api.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/github/intellectronica/agent-skills/notion-api.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow