Back to skill

Security audit

Jquants Mcp

Security checks for vulnerabilities and agentic risk

Overview

The skill is a straightforward J-Quants market-data helper, with the main caution that it installs an unpinned third-party command that uses J-Quants credentials.

Before installing, confirm you trust the `jquants-mcp` package source and consider pinning a reviewed version. Use only the J-Quants credentials needed for this account, avoid exposing unrelated secrets in the same environment, and rotate the password if you suspect the installed package was compromised.

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:4
Finding
Unpinned Third-Party Package Receives Account Credentials## Vulnerability Details **File Location**: `SKILL.md`, line 4 and lines 76-78 **Vulnerability Type**: `T08: Insecure Dependencies` **Risk Level**: Medium The skill installs the `jquants-mcp` package without pinning a version, verifying a package hash or signature, or identifying a specific audited source revision. The resulting executable runs in an environment containing the user's J-Quants email address and password. **Relevant code at line 4:** ```yaml metadata: {"openclaw":{"emoji":"💹","requires":{"bins":["jquants-mcp"],"env":["JQUANTS_MAIL_ADDRESS","JQUANTS_PASSWORD"]},"install":[{"id":"uv","kind":"uv","package":"jquants-mcp","bins":["jquants-mcp"],"label":"Install jquants-mcp (uv)"}],"tags":["japan","stock","jpx","tse","price","ohlcv","finance","mcp","jquants"]}} ``` **Relevant code at lines 76-78:** ```markdown - Requires `JQUANTS_MAIL_ADDRESS` and `JQUANTS_PASSWORD` environment variables - Free account registration: https://jpx-jquants.com/ - Python package: `pip install jquants-mcp` or `uv tool install jquants-mcp` ``` ### Technical Analysis Both documented installation methods resolve the package without a fixed version or integrity constraint. Consequently, the code installed can change between installations even when the reviewed skill remains unchanged. No package source, lockfile, cryptographic hash, signature, or audited commit is included in the project. The installed command requires access to `JQUANTS_MAIL_ADDRESS` and `JQUANTS_PASSWORD`. A malicious or compromised package release could read these environment variables during execution and transmit them externally. It could also execute arbitrary code with the permissions of the user invoking the command. The inspected project does not establish that the current package is malicious; the vulnerability is the uncontrolled supply-chain trust boundary and direct exposure of credentials to unverifiable dependency code. ### Attack Path 1. An attacker c ...[truncated 1324 chars]
Remediation
## Remediation Suggestions 1. Pin `jquants-mcp` to a specific, reviewed version rather than resolving the latest available release. 2. Verify the package using a cryptographic hash or trusted signature and maintain those integrity values in a lockfile or installation manifest. 3. Document the authoritative publisher and source repository, and reference a reviewed release or immutable commit. 4. Include auditable source code in the skill package where practical, or perform a separate security review of the exact dependency version before deployment. 5. Prefer a scoped, revocable API token over an account password if J-Quants supports one. 6. Execute the dependency in a restricted environment with only the required credentials, filesystem access, and network permissions. 7. Avoid exposing unrelated secrets through the process environment, and rotate the J-Quants password immediately if dependency compromise is suspected.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Rp1

Low
Category
MCP Rug Pull
Confidence
95% confidence
Finding
The skill instructs users to install `jquants-mcp` without pinning an exact version, which makes the installed code non-reproducible and exposes users to supply-chain risk if a later package release is compromised or introduces breaking behavior. In this context, the package is an executable MCP-related tool and the skill also requires credentials via environment variables, so an unsafe future version could access sensitive J-Quants account data or misuse the host environment.

Static analysis

No suspicious patterns detected.