Evolver 1.32.2

• Outcome: Review
• Latest audit: Sep 16, 2026 · 8:54 AM UTC
• Version: 1.0.0

Overview

This skill is a disclosed self-evolution tool, but it gives remote Hub content and automated loops too much influence over code changes and data sharing.

Install only in a disposable or tightly controlled repository. Use review mode, set EVOLVE_BRIDGE=false unless you explicitly want spawned executor behavior, prefer EVOLVER_ROLLBACK_MODE=stash, avoid providing broad GitHub tokens, and use only a trusted HTTPS A2A_HUB_URL. Do not run loop mode around private transcripts or secrets unless outbound Hub sharing is acceptable.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Insecure Skill Coding Practices: Finds exploitable flaws such as hardcoded secrets or command injection
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally

Findings (3)

T01 · Skill Instruction Hijacking

Error

• Location: src/gep/prompt.js:196

Finding

Untrusted Hub Content Reaches an Execute-by-Default Agent Prompt

Content

View full analysis

0) { result.relevant_lessons = respPayload.relevant_lessons; } return result; ``` The remote lessons are retained for prompt generation: ```js // LessonL: capture relevant lessons from Hub if (Array.isArray(fetchResult.relevant_lessons) && fetchResult.relevant_lessons.length > 0) { hubLessons = fetchResult.relevant_lessons; console.log(`[LessonBank] Received ${hubLessons.length} lesson(s) from ecosystem.`); } ``` Remote content is represented as trusted cross-agent experience and the executor is instructed to apply it: ```js function buildLessonsBlock(hubLessons, signals) { if (!Array.isArray(hubLessons) || hubLessons.length === 0) return ''; var sigSet = new Set((Array.isArray(signals) ? signals : []).map(function (s) { return String(s).toLowerCase(); })); var positive = []; var negative = []; for (var i = 0; i < hubLessons.length && (positive.length + negative.length) < 6; i++) { var l = hubLessons[i]; if (!l || !l.content) continue; var entry = ' - [' + (l.scenario || l.lesson_type || '?') + '] ' + String(l.content).slice(0, 300); if (l.source_node_id) entry += ' (from: ' + String(l.source_node_id).slice( ...[truncated 5426 chars]

Remediation

View remediation

T09 · Insecure Skill Coding Practices

Error

• Location: src/gep/questionGenerator.js:97

Finding

Raw User Transcript Excerpts May Be Sent to the EvoMap Hub

Content

View full analysis

0) { var featureContext = featureLines[0].replace(/\s+/g, ' ').trim().slice(0, 150); candidates.push({ question: 'User requested a feature that may benefit from community solutions: ' + featureContext + ' -- Are there existing implementations or best practices for this?', amount: 0, signals: ['user_feature_request', 'community_solution_sought'], priority: 1, ...[truncated 3648 chars]

Remediation

View remediation

other

Warning

• Location: src/gep/envFingerprint.js:42

Finding

Stable Device Fingerprint and Host Telemetry Are Sent to a Configurable Hub Endpoint

Content

View full analysis

Remediation

View remediation

SkillSpector

Vulnerability Patterns

• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (190)

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 97% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 92% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 96% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 97% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 98% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 97% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 97% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 96% confidence

Finding

The skill presents itself as a self-evolution engine but the documented behavior spans repo manipulation, remote hub communication, issue creation, asset ingestion, task handling, and process management. That breadth materially increases the effective privilege and trust required, making operator misinformed consent and unexpected high-risk actions more likely, especially given shell and network permissions plus autonomous default execution.

Content

No source excerpt is available for this finding.