Back to skill

Security audit

OpenServ Launch

Security checks for vulnerabilities and agentic risk

Overview

The skill does what it says, but it can trigger real Base mainnet token launches without a clear confirmation or risk warning.

Review this carefully before installing or running examples. Only use it if you intend to create public Base mainnet tokens through OpenServ, verify the exact token parameters and wallet address before every launch, avoid running the provided launch script as-is, and prefer pinned package versions for npx and skill update commands.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:217
Finding
Unpinned Packages Executed Through npx## Vulnerability Details **File Location**: `SKILL.md`, lines 217-226 **Vulnerability Type**: Unpinned executable dependency and mutable Skill update source **Risk Level**: Medium ```bash # Check if updates are available npx skills check # Update all installed skills to latest versions npx skills update ``` ```bash npx skills add openserv-labs/skills ``` ### Technical Analysis The documented commands invoke the unversioned `skills` npm package through `npx`. If the package is not already installed locally, `npx` can resolve, download, and execute the current registry version under the user's account. The instructions do not pin an exact package version, verify a package integrity hash, or identify a reviewed Skill commit. The update and installation commands also retrieve mutable Skill content. Consequently, the code and instructions that ultimately execute can differ from the content reviewed in this audit. This creates a supply-chain trust boundary in which compromise of the package, publisher account, package registry, or upstream Skill repository could introduce attacker-controlled behavior. ### Attack Path 1. An attacker compromises the relevant npm package, publisher credentials, registry resolution path, or upstream Skill source. 2. The attacker publishes or serves a malicious version while preserving the expected package or repository name. 3. A user follows the instructions and runs one of the unversioned `npx skills` commands. 4. `npx` resolves and executes the mutable package under the user's local privileges. 5. The malicious package can access resources available to that user. Alternatively, an update can install malicious Skill instructions that influence later agent operations. ### Impact Assessment A malicious npm payload would execute with the privileges of the user running `npx`. Depending on the execution environment, this could expose accessible files, environment variables, development ...[truncated 379 chars]
Remediation
## Remediation Suggestions 1. Pin the CLI to an exact reviewed version, for example `npx --yes skills@X.Y.Z`, after verifying the correct package identity. 2. Use a lockfile and require npm integrity verification for executable dependencies. 3. Pin Skill installation to an immutable commit hash or signed release rather than a mutable repository head. 4. Document the expected npm scope, publisher, repository, release checksum, and signature-verification procedure. 5. Separate checking from applying updates; display the proposed version and changes, then require explicit review before installation. 6. Run package and Skill installation in a restricted environment without production secrets or elevated privileges. 7. Re-audit updated Skill content before trusting or executing it.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (16)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding
The declared description centers on token creation and launch operations on Base blockchain through the OpenServ Launch API. However, the supplied code chunk only performs GET requests to /api/tokens and /api/tokens/{address} to retrieve token metadata. It has no code for deploying contracts, launching tokens, creating liquidity pools, or initiating blockchain transactions. This is a materially different primary behavior: discovery and inspection of launched tokens rather than token launch. Therefore, the description does not accurately represent this code chunk.

Credential Access

High
Category
Privilege Escalation
Content
*   npx tsx agent-launcher.ts
 *
 * Prerequisites:
 *   - OPENAI_API_KEY in .env
 */

import 'dotenv/config'
Confidence
60% confidence
Finding
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill contains executable example code that performs network actions and references environment capability, but the manifest does not declare any explicit tool scope or allowed-tools boundary. In an agent ecosystem, missing permission declarations can lead to over-broad execution privileges or unclear review expectations, increasing the chance that an agent can access environment data or invoke tooling beyond what a user intended.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill promotes irreversible on-chain token deployment and liquidity actions without prominently warning about financial cost, immutability, and the inability to undo transactions once submitted. In the context of blockchain operations, omission of these warnings increases the risk of accidental loss, unintended token launches, or users authorizing expensive actions without informed consent.

Rp1

Medium
Category
MCP Rug Pull
Confidence
83% confidence
Finding
The skill instructs users to run `npx skills check` without pinning a version, which can fetch and execute whatever package version is current at execution time. This creates a supply-chain risk because a compromised or breaking upstream release could execute unexpected code on the user's system.

Rp1

Medium
Category
MCP Rug Pull
Confidence
83% confidence
Finding
The unpinned `npx skills update` command executes the latest published package version at runtime, exposing users to supply-chain compromise or unintended behavior changes. Because this command is explicitly about updating installed skills, it is especially sensitive to trust in upstream package distribution.

Rp1

Medium
Category
MCP Rug Pull
Confidence
84% confidence
Finding
The reinstall instruction `npx skills add openserv-labs/skills` is unpinned and may execute a different package version over time, creating avoidable supply-chain exposure. Users following documentation may run arbitrary newly-published code with local execution privileges.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The `launch_token` capability triggers a real external token-creation API call that can cause irreversible on-chain or platform-side actions, yet the agent does not enforce an explicit confirmation or present a clear warning immediately before execution. In an LLM-agent context, this increases the chance of accidental or prompt-induced token launches, especially because the skill is specifically designed to perform high-impact blockchain actions.

Rp1

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding
The usage instruction `npx tsx launch-token.ts` executes a package resolver path without pinning an exact version, which can cause users to run an unexpected or recently compromised `tsx` release. In a blockchain token-launching context, that increases risk because the script handles wallet-related configuration and triggers external launch actions, so a malicious dependency version could alter behavior, exfiltrate environment data, or execute arbitrary code on the operator's machine.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The reference documents an endpoint that creates tokens, mints liquidity positions, locks LP, and triggers on-chain transactions, but it does not prominently warn that these actions may be irreversible, consume gas, and create real financial/asset consequences. In an agent-skill context, missing a transaction-risk warning can cause users or downstream agents to invoke the launch flow without informed consent or adequate confirmation.

External Transmission

Medium
Category
Data Exfiltration
Content
### Example Request

```bash
curl -X POST "https://instant-launch.openserv.ai/api/launch" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "My Token",
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
### Example Request

```bash
curl -X POST "https://instant-launch.openserv.ai/api/launch" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "My Token",
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.