Back to skill

Security audit

meta-game-negotiation

Security checks for vulnerabilities and agentic risk

Overview

This skill is not malicious, but it claims broad self-evolving agent behavior while shipping only a small persistent learner that stores arbitrary notes without clear controls.

Review this before installing if you expect a complete negotiation tool or do not want cross-session learning. Use it only where local persistent notes are acceptable, keep learned_patterns.json under user control, and ensure any meta-evolver or reflection system never treats stored notes as instructions without human review.

Vulnerability Patterns
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Findings (1)

T02 · Agent Memory Poisoning

Warning
Location
scripts/learner.py:10
Finding

Unvalidated Attacker-Controlled Notes Written to Persistent Agent 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 an arbitrary command-line string and appends it to the persistent `patterns` collection in `scripts/learned_patterns.json`. It performs no schema validation, provenance tracking, trust classification, content filtering, length restriction, or record-count limitation. The skill documentation states that execution results and failure patterns are intended to enter a self-reflection and cross-session evolution workflow. Consequently, a task author or other party able to influence the note argument can persist instruction-like content or false observations in the state expected to inform future agent activity. The repository does not contain the downstream meta-evolver or memory-consumption implementation. Therefore, direct alteration of future model behavior is conditional on an external agent subsequently loading and treating these stored strings as trusted guidance. The persistent write itself is present and confirmed. ### Attack Path 1. An attacker supplies content that is propagated into the `note` argument, such as a false operational rule or instruction-like text. 2. The skill or an orchest ...[truncated 1373 chars]
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 (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The declared description presents a highly capable meta-skill with advanced autonomous reasoning and evolution features. The supplied code chunk only implements a lightweight local persistence mechanism for recording successes, failures, and notes in a JSON file. While this could be a supporting component for a larger meta-evolution loop, on its own it does not substantiate the primary declared purpose and materially understates the gap between description and actual behavior. There are no suspicious extra permissions or unrelated external accesses beyond local file I/O, but the description significantly overstates the implemented capability.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill advertises capabilities related to persistence and reflective loops, and the static analyzer detected file-write behavior, but the manifest does not declare any tool scope such as permissions or allowed-tools. Undeclared write capability weakens least-privilege controls and makes it harder for a host agent or reviewer to understand what side effects the skill may have, especially because the skill also claims long-term memory and self-evolution behavior.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The natural-language content of the skill description and usage instructions is presented entirely in Chinese, with no indication that users may opt into another language or that the skill is intentionally region-specific. This can violate language/locale policy because it implicitly constrains interaction language for all users.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This markdown file is entirely written in Chinese and presents the report content as fixed-language output, with no indication that the user can choose another language or opt in to this locale. Under the language/locale policy rule, fixed language behavior without user choice is a natural-language policy concern unless clearly justified as region-specific.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

This code persists operation results and optional notes to a local JSON file, which is a file write affecting user or skill data. Although the module docstring hints at recording calls, there is no confirmation prompt, print/log disclosure, or inline warning at the write site explaining that notes will be stored on disk.

Content

No source excerpt is available for this finding.