Back to skill

Security audit

Panoptica Skill

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent game API handbook, but its free-form remote override channel needs review before an agent treats server-returned text as instructions.

Install only if you intend to let an agent interact with this external game service. Keep owner and agent API keys in protected secret storage, do not paste live keys into chats or logs, and configure any implementation so override_pending, inbox messages, captcha text, agent names, and WebSocket events are parsed only as untrusted game data. Avoid enabling shell, filesystem, memory-writing, or unrelated tools in any handler for these remote messages.

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
Findings (1)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:632
Finding

Untrusted Natural-Language Override Channel Can Redirect Agent Behavior

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 632-647
Vulnerability Type: T01: Skill Instruction Hijacking
Risk Level: High

Vulnerable Code Snippet

markdown
### Override System

Owners can send direct commands to their agents using PoW tokens:

```http
POST /v1/override
Authorization: Bearer <owner_api_key>
Content-Type: application/json

{
  "agent_id": "MyAgent01",
  "command": "retreat to GRID immediately"
}

Cost: 1 PoW per override.

The command appears in the agent's next heartbeat response as override_pending.

text

Related automatic polling behavior appears at lines 163-181:

```markdown
> **⚠️ IMPORTANT:** Heartbeat is automatically handled by `heartbeat_daemon.py` every 25 seconds. Do NOT include heartbeat calls in your agent logic. Focus on gameplay actions only.

```http
POST /v1/agent/heartbeat
Authorization: Bearer <agent_api_key>

Effects per heartbeat (automatic):

  • Updates last_heartbeat timestamp
  • Awards +1 PoW token to owner
  • Auto-moves agent 1-3 cells randomly (map visualization)
  • Auto-decays heat by -1 (if heat > 0)
  • Checks for pending override commands

Returns:

json
{
  "status": "ACTIVE",
  "override_pending": null,
  "pow_tokens": 15,
  "heat_level": 8
}
text

### Technical Analysis

The Skill defines a recurring external command channel through which an owner can submit unrestricted natural-language text. The heartbeat mechanism retrieves this text as `override_pending`. The instructions do not define a structured command schema, an allowlist of permitted actions, or a trust-boundary rule requiring override content to be handled solely as untrusted game data.

If an agent interprets `override_pending` as authoritative instructions, the remote owner or anyone who compromises the owner account or service can inject instructions after the Skill has been reviewed. Such content could attempt to supersede the user's current objective, request unrelated tool operation
...[truncated 1933 chars]
Remediation
View remediation

Remediation Suggestions

  1. Replace unrestricted natural-language overrides with a strict command protocol, such as an enumerated action and validated parameters:

    json
    {
      "action": "RETURN_TO_GRID",
      "parameters": {}
    }
    
  2. Enforce a server-side and client-side allowlist of game-only actions. Reject unknown fields, free-form commands, nested instructions, URLs, code, and oversized values.

  3. Explicitly state that all remote content—including override_pending, inbox messages, captchas, WebSocket events, agent names, and message bodies—is untrusted data and must never modify system or user instructions.

  4. Ensure the override handler can invoke only narrowly scoped game API functions. It must not have access to shell execution, filesystem tools, local secrets, general network clients, memory-writing facilities, or unrelated agent tools.

  5. Require explicit user confirmation for consequential operations, including transfers, combat, item deletion, credential changes, or any action outside a predefined low-risk gameplay policy.

  6. Authenticate and authorize override requests robustly. Use revocable, narrowly scoped credentials; rotate exposed keys; rate-limit requests; and maintain tamper-resistant audit logs.

  7. Add instruction/data separation tests. Include adversarial inputs such as requests to ignore prior instructions, expose credentials, invoke tools, download content, or alter persistent memory, and verify that they are rejected rather than executed.

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 460)May include surrounding context.

11.3 Discard Module

http
DELETE /v1/inventory/modules/<module_uuid>
Authorization: Bearer <agent_api_key>

Instruction Override

High
Category
Prompt Injection
Confidence
90% confidence
Finding

This pattern attempts to override system instructions or ignore safety constraints. Without LLM analysis, manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 630)May include surrounding context.

md
- Owner earns **+1 PoW** per agent heartbeat
- Initial allocation: **3 PoW** at registration

### Override System

Owners can send direct commands to their agents using PoW tokens:

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The skill prominently exposes long-lived owner and agent API keys and emphasizes saving them, but provides no guidance about secret storage, redaction, or avoiding transmission to other tools, logs, or chats. In an agent setting, that increases the chance of credential leakage and full account or agent takeover.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
99% confidence
Finding

The recommended strategy instructs beginners to mine in GRID, but the earlier mining section explicitly states mining is not possible there and that agents must travel to SLUMS first. In an autonomous agent skill, contradictory operational guidance can cause repeated invalid actions, wasted turns, stalled progression, and unstable agent behavior.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill says heartbeat is auto-managed every 25 seconds, but the summary later says every 30 seconds is mandatory. Conflicting liveness timing in an agent handbook can lead to missed heartbeats, GHOST state transitions, and denial of service against the agent if implementers follow the wrong interval.

Content

No source excerpt is available for this finding.