Security audit

SiteSpeakAI

Security checks for vulnerabilities and agentic risk

Overview

This SiteSpeakAI skill is mostly purpose-aligned, but it gives broad connector access to sensitive account data and includes unsafe one-line remote installer commands.

Review before installing. Use safer manual installation for the oo CLI, verify the installer or package source, and only connect a SiteSpeakAI account with the minimum permissions needed. Ask the agent to confirm scope before retrieving conversations, leads, account details, or broad chatbot data.

A.I.G

Vulnerability Patterns

Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
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
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally

Findings (2)

T03 · Remote Payload Retrieval and Execution

Error

Location: SKILL.md:64
Finding: Unverified Remote Installation Scripts Are Executed Directly
Content: View full analysis
Remediation: View remediation

T05 · Unauthorized Access and Privilege Escalation

Warning

Location: SKILL.md:22
Finding: Broad Connector Access Can Expose Sensitive SiteSpeakAI Records
Content: View full analysis
" ``` ```bash oo connector run "sitespeakai" --action "" --data '' --json ``` ```text - `list_conversations` — Retrieve conversation history for one SiteSpeakAI chatbot. - `list_leads` — List the leads captured by one SiteSpeakAI chatbot. ``` ### Technical Analysis The Skill directs every SiteSpeakAI request through the OOMOL connector and relies on credentials injected by OOMOL. Available read operations include complete conversation histories and captured leads, which may contain personal, confidential, or customer-provided information. The use of OOMOL and server-side credentials is disclosed, and the reviewed file contains no evidence of covert exfiltration. However, the blanket instruction to use the connector for any SiteSpeakAI request, combined with the policy that untagged read actions may run directly, can permit retrieval of sensitive datasets without an additional confirmation or a documented data-minimization step. The connector legitimately requires network access to provide the declared functionality. The risk arises from potentially broader account and record access than a particular user request requires, rather than from network communication alone. ### Attack Path 1. A user makes a SiteSpeakAI-related request, or an ambiguous request is interpreted as requiring SiteSpeakAI data. 2. The Skill selects an untagged read action such as `list_conversations` or `list_leads`. 3. OOMOL uses the con ...[truncated 1002 chars]
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

External Script Fetching

High

Category: Supply Chain
Confidence: 98% confidence
Finding: The skill instructs users to install software by piping a remote script directly into bash, which executes network-fetched code without prior verification. If the install endpoint, transport path, or hosting account were compromised, this could lead to arbitrary code execution on the user's machine.
Content: Scanner excerpt · SKILL.md (reported line 65)May include surrounding context.

oo: command not found — install the oo CLI (other platforms: https://cli.oomol.com/install-guide.md):

bash
curl -fsSL https://cli.oomol.com/install.sh | bash # macOS / Linux

powershell

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger description instructs the agent to use this skill for ANY SiteSpeakAI request, which is broader than necessary and can cause automatic invocation whenever SiteSpeakAI is merely mentioned. In an agentic setting, this increases the chance of unintended access to connected account data or accidental execution of write/destructive actions without sufficient task disambiguation.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.