Back to skill

Security audit

meta-fintech-engineer

Security checks for vulnerabilities and agentic risk

Overview

The skill is not destructive, but its claimed fintech/meta-agent abilities are unsupported and it persists arbitrary learning notes without clear boundaries.

Review before installing. Treat this as an experimental generated Chinese-language learner stub, not a complete fintech engineering skill. Reset or remove the packaged learned_patterns.json, avoid feeding its stored notes directly into agent instructions, and require human review before any learned pattern becomes future behavior.

Vulnerability Patterns
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • 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
Findings (2)

T02 · Agent Memory Poisoning

Warning
Location
scripts/learner.py:5
Finding

Unsanitized Attacker-Controlled Data Persisted as Cross-Session Learning State

Content
View full analysis
1 else True rec = (sys.argv[2] if len(sys.argv) > 2 else "") print(json.dumps(record(ok, rec), ensure_ascii=False)) ``` ### Technical Analysis The learner accepts its second command-line argument as an arbitrary note and appends it directly to the persistent `patterns` array. It applies no schema validation, provenance tracking, trust classification, content sanitization, length restriction, or entry-count limit. The Skill documentation describes this file as supporting cross-session learning and continuous evolution. If a subsequent reflection or meta-evolver component consumes these entries as behavioral guidance, an attacker can store instruction-like content that may influence future sessions. The implementation also fails to distinguish untrusted observations from trusted rules. The included project does not contain the downstream component that interprets these patterns. Therefore, direct instruction execution is not demonstrated within the audited package, but the persistent untrusted-data channel and its intended use as learning state are present. ### Attack Path 1. An attacker supplies crafted text through task content or another input that reaches the learner invocation. ...[truncated 1263 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Note
Location
scripts/learned_patterns.json:5
Finding

Packaged Runtime State Discloses a Local User and Workstation Path

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (6)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The declared description presents a broad, advanced meta-agent with layered capabilities such as self-verification, reflection, orchestration, and continuous self-improvement. The supplied code chunk only implements a minimal telemetry/persistence utility: it loads a JSON file, increments total operation and failure counters, optionally appends notes, and writes the results back. While this could support a larger evolution loop as a helper component, on its own it does not substantiate the claimed primary purpose. Therefore the description materially overstates what the code actually does.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
83% confidence
Finding

The skill advertises behaviors such as self-reflection and writing to learner/self-reflection loops, which implies state or file modification capability, but it does not declare any explicit tool scope or permissions. Undeclared write capability weakens reviewability and least-privilege controls, making it easier for a skill to persist data or modify local state in ways operators did not authorize.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The primary natural-language description is written entirely in Chinese, and the document provides no option for user language selection or note that the skill is intentionally region- or locale-specific. Under the stated policy, forcing a specific language without opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The only natural-language description in the file is written in Chinese, which can impose a language/locale expectation without offering user opt-in or documenting why that locale is required. Under the policy, language constraints should either be optional for the user or clearly justified as region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

This markdown file contains user-facing natural language only in Chinese, and it does not state that the skill is region-specific or that users can opt into another language. Under the language/locale policy rule, forcing a specific language without user choice can be a policy concern.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

This code writes persistent data to learned_patterns.json, which affects local user/system data. Although the module docstring states it records skill invocation results, there is no confirmation prompt, user-facing log message, or inline warning at the write site to disclose that persistence is happening.

Content

No source excerpt is available for this finding.