ClawHub
Back to skill

Security audit

sentisense

Security checks for vulnerabilities and agentic risk

Overview

The skill’s market-data access, temporary-key setup, and optional watchlist queries are disclosed and proportionate to its research purpose.

SentiSense receives your requested tickers, screening filters, and any watchlist you submit. Without an existing key, the agent is instructed to create a free temporary one automatically. Keep the key and its keep link private. Use the REST path if you want to avoid optional package execution and CLI configuration.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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 (4)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The unconditional 'no write operations' claim conflicts with instructions to create a temporary API credential automatically when none is available. This is a limited authorization-boundary mismatch: an agent relying on the declared read-only scope could create provider-side state without separate user approval, although the provisioning behavior is openly documented and does not authorize trading or purchases.

Content

No source excerpt is available for this finding.

The skill automatically provisions a new API credential despite its manifest's explicit read-only, no-write guarantee.

Medium
Category
Excessive Agency
Confidence
95% confidence
Finding

The instruction to 'get a temporary one yourself' initiates credential creation rather than merely retrieving financial data, despite the manifest's no-write guarantee. Temporary expiry, request limits, closed account endpoints, and explicit secret-handling guidance reduce the consequences; this is the same underlying scope mismatch as TP4, not an additional independent vulnerability.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 1677)May include surrounding context.

| limit | int | No | 100 | Rows returned, capped at 500. Sits here, not inside plan |

bash
curl -X POST "https://app.sentisense.ai/api/v1/screener/execute" \
  -H "X-SentiSense-API-Key: $SENTISENSE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{

The watchlist-screening instructions omit a privacy warning about transmitting the user's watchlist to SentiSense.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The skill instructs the agent to include a watchlist in an authenticated request to the third-party SentiSense API. Although network access is documented, there is no warning that this shares the user's financial interests with the provider and associates them with the API key.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.