Admapix Ice

WarnAudited by ClawScan on May 18, 2026.

Overview

The skill’s ad analytics purpose is coherent, but its deep-research path forwards your AdMapix API key to a separate service and embeds user text in shell commands in a way that needs review.

Review the deep-research behavior before installing. Basic AdMapix API use appears purpose-aligned, but deep queries may send your API key and business context to deepresearch.admapix.com and run shell commands built from user text. Prefer using it only with non-sensitive prompts and a scoped/revocable API key.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

A crafted prompt could cause the agent to run an unintended shell command or send unintended data while using the user’s API key.

Why it was flagged

The user’s query and context are inserted into a shell command inside a JSON string. If not safely escaped, user-controlled text containing quotes or shell syntax could alter the command or request body.

Skill content
Run this exact command (only replace `{user_query}` and `{additional_context}`): ... -d '{"project": "admapix", "query": "{user_query}", "context": "{additional_context}", "api_key": "'"$ADMAPIX_API_KEY"'"}'
Recommendation

Use a safe HTTP client or structured API tool that JSON-encodes fields, avoid raw shell interpolation, and require confirmation before running deep-research commands.

What this means

Your AdMapix API key is delegated to another AdMapix-hosted service for deep analysis, with no clear limits, retention policy, or per-task consent described.

Why it was flagged

The skill instructs the agent to forward the user’s AdMapix API key to the deep-research framework, rather than only using it directly against the declared api.admapix.com API.

Skill content
`api_key` passes the user's API key to the framework — always include it as shown above.
Recommendation

Only use the deep-research path if you trust that service with your API key; the publisher should document key handling, avoid forwarding raw keys where possible, or use scoped short-lived tokens.

What this means

Potentially sensitive business questions or context may be sent to a remote research service whose data handling is not explained in the artifacts.

Why it was flagged

The deep-research workflow sends the user’s query, optional business context, and API key to a separate remote task service using a hardcoded bearer token, with unclear identity, permission, and data-boundary controls.

Skill content
curl -s -X POST "https://deepresearch.admapix.com/research" ... -H "Authorization: Bearer test-local-token-2026" ... "query": "{user_query}", "context": "{additional_context}"
Recommendation

Avoid including confidential business details unless you trust the provider; the skill should disclose the remote service, authentication model, data retention, and whether submitted context is stored.

What this means

It may be harder to verify exactly which publisher/version produced the reviewed skill artifacts.

Why it was flagged

The bundled metadata differs from the supplied registry metadata, which lists a different owner ID and version. This does not prove unsafe behavior, but it is a provenance inconsistency.

Skill content
"ownerId": "kn7c1c01gzrc3m423t8n840m9s81vj6m", "slug": "admapix-ice", "version": "1.0.26"
Recommendation

Confirm the publisher and version before installing, especially because the skill handles an API key.