ClawHub
Back to skill

Security audit

pinterest-research

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a Pinterest research helper, but its bundled API script is broader than the stated Pinterest-only purpose and can send the API key to non-Pinterest or overridden destinations.

Review before installing. This does not show destructive behavior or hidden persistence, but users should understand that the helper is not limited to Pinterest and should avoid setting CRAWLORA_API_BASE unless they fully trust the destination. A safer version would restrict the script to documented /pinterest endpoints and pin the trusted API base.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The helper is documented as a generic Crawlora client and explicitly advertises non-Pinterest endpoints such as Amazon, YouTube, Google Search, and Google Trends. In a skill whose declared purpose is Pinterest-only research, this creates a scope-expansion vulnerability: callers can repurpose the skill to access unrelated data sources and capabilities through the same privileged API key.

Description-Behavior Mismatch

High
Confidence
99% confidence
Finding
The script accepts an arbitrary path from the caller and concatenates it directly onto the Crawlora API base, allowing access to any Crawlora endpoint the API key can reach. Because this skill is supposed to be limited to Pinterest research, the unrestricted path parameter turns it into a generic outbound API proxy and materially increases the risk of policy bypass, unintended data access, and misuse of the provided credentials.

Context-Inappropriate Capability

Medium
Confidence
93% confidence
Finding
Allowing the API base URL to be overridden via CRAWLORA_API_BASE lets an attacker or misconfigured environment redirect authenticated requests, including the x-api-key header and request bodies, to an arbitrary host. In this skill context, there is no clear Pinterest-specific need for dynamic base override, so it meaningfully increases the chance of credential exfiltration and unauthorized external transmission.

External Transmission

Medium
Category
Data Exfiltration
Content
curl -fsS -G "${auth[@]}" "${qs[@]}" "${base}${path}"
else
  [ -n "$body" ] || body="${rest[0]:-{}}"
  curl -fsS -X "$method" "${auth[@]}" \
    -H "Content-Type: application/json" -d "$body" "${base}${path}"
fi
Confidence
88% confidence
Finding
curl -fsS -X "$method" "${auth[@]}" \ -H "Content-Type: application/json" -d

External Transmission

Medium
Category
Data Exfiltration
Content
set -euo pipefail

: "${CRAWLORA_API_KEY:?Set CRAWLORA_API_KEY first — get a free key at https://crawlora.net?utm_source=github&utm_medium=referral&utm_campaign=crawlora-skills}"
base="${CRAWLORA_API_BASE:-https://api.crawlora.net/api/v1}"

method="GET"
body=""
Confidence
90% confidence
Finding
https://api.crawlora.net/

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
SKILL.md:25