Back to skill

Security audit

jpverify

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marked this skill suspicious because scanner results report invoice uploads that can send complete documents to an external service without a clear privacy warning, as well as capabilities not disclosed in its description.

Review what invoice data would be sent to JPVerify and whether that sharing is acceptable under your data-sharing rules before installing or using the invoice-upload examples. Also consider whether you need the undisclosed filings and regulated-business checks; the scanner's partial coverage means evidence is incomplete.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (89)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The declared description lists specific verification, address, company-name resolution, and invoice-validation capabilities, but does not disclose corporate-filings retrieval. The supplied code performs that distinct retrieval operation rather than one of the listed checks. Authentication and use of the JPVerify API are consistent supporting details and are not themselves mismatches.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
89% confidence
Finding

Although corporate filings concern Japanese business data and the request uses the expected JPVerify service, retrieving filings is a distinct capability not disclosed in the description. The supplied chunk does not show this retrieval supporting any declared verification operation. Authentication and client-version headers are ordinary implementation details, not additional mismatches.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding

The code requests a category-specific regulatory-status check, which is distinct from the declared corporate-number, invoice, supplier, address and e-invoice checks. Although it uses the expected JPVerify service and ordinary API authentication, the description does not disclose regulated-entity verification.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
87% confidence
Finding

The code performs a category-specific financial regulatory check, rather than the company identity, invoice, address, or document checks described. Although related to Japanese business verification, regulated-business status verification is a distinct capability not disclosed in the description. API-key access and the request to JPVerify are consistent supporting implementation details, not separate mismatches.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 34)May include surrounding context.

md
1. **API key (recommended).** If the environment has `JPVERIFY_API_KEY` (a `jpv_live_…` key the user created), send it on every call:
   - HTTP: header `Authorization: Bearer $JPVERIFY_API_KEY` (or `X-API-Key`).
   - MCP: if the JPVerify Claude Code plugin is installed, the `jpverify` MCP server is already configured and uses the key saved in the plugin's settings (`/plugin` → JPVerify → Configure), not `JPVERIFY_API_KEY`. Otherwise: `claude mcp add --transport http jpverify https://api.jpverify.com/mcp --header "Authorization: Bearer $JPVERIFY_API_KEY" --header "X-JPVerify-Client: jpverify/1.3.5"`
   Each successful call is deducted from the user's prepaid USD credits; failed or invalid calls are refunded automatically. Before a large batch, check the balance for free: `GET https://api.jpverify.com/v1/account/balance` with the same header, or the MCP tool `jpverify_balance`.
2. **x402 wallet.** If you have an x402 client with USDC on Base (`eip155:8453`) and the user approved a budge

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 35)May include surrounding context.

md
MCP server is already configured and uses the key saved in the plugin's settings (`/plugin` → JPVerify → Configure), not `JPVERIFY_API_KEY`. Otherwise: `claude mcp add --transport http jpverify https://api.jpverify.com/mcp --header "Authorization: Bearer $JPVERIFY_API_KEY" --header "X-JPVerify-Client: jpverify/1.3.5"`
   Each successful call is deducted from the user's prepaid USD credits; failed or invalid calls are refunded automatically. Before a large batch, check the balance for free: `GET https://api.jpverify.com/v1/account/balance` with the same header, or the MCP tool `jpverify_balance`.
2. **x402 wallet.** If you have an x402 client with USDC on Base (`eip155:8453`) and the user approved a budget, call without a key: the API answers HTTP 402 with the exact price; sign and retry with `PAYMENT-SIGNATURE`. Pay only within the approved budget.
3. **Neither.** Do not pay and do not invent a key. The MCP server also works without a key for the free tool `jpverify_guide` (`claude mcp

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 37)May include surrounding context.

md
deducted from the user's prepaid USD credits; failed or invalid calls are refunded automatically. Before a large batch, check the balance for free: `GET https://api.jpverify.com/v1/account/balance` with the same header, or the MCP tool `jpverify_balance`.
2. **x402 wallet.** If you have an x402 client with USDC on Base (`eip155:8453`) and the user approved a budget, call without a key: the API answers HTTP 402 with the exact price; sign and retry with `PAYMENT-SIGNATURE`. Pay only within the approved budget.
3. **Neither.** Do not pay and do not invent a key. The MCP server also works without a key for the free tool `jpverify_guide` (`claude mcp add --transport http jpverify https://api.jpverify.com/mcp`); use it to confirm which call fits, then tell the user they can create a free account with starter credit at https://www.jpverify.com/signup?ref=jpverify and set `JPVERIFY_API_KEY`.

Always send `X-JPVerify-Client: jpverify/1.3.5` on HTTP calls. Never print, log or share the API key.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 143)May include surrounding context.

API key:

bash
curl -X POST https://api.jpverify.com/v1/jp/companies/resolve \
  -H "Authorization: Bearer $JPVERIFY_API_KEY" \
  -H "X-JPVerify-Client: jpverify/1.3.5" \
  -H "Content-Type: application/json" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 154)May include surrounding context.

md
import os, requests

response = requests.post(
    "https://api.jpverify.com/v1/jp/companies/resolve",
    headers={"Authorization": f"Bearer {os.environ['JPVERIFY_API_KEY']}",
             "X-JPVerify-Client": "jpverify/1.3.5"},
    json={"name": "Murata Manufacturing Co., Ltd.", "prefecture": "Kyoto"},

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 167)May include surrounding context.

bash
# No key: shows the HTTP 402 x402 price; pay with an x402 client and retry with PAYMENT-SIGNATURE.
curl -i -X POST https://api.jpverify.com/v1/jp/companies/resolve \
  -H "X-JPVerify-Client: jpverify/1.3.5" \
  -H "Content-Type: application/json" \
  -d '{"name": "Murata Manufacturing Co., Ltd.", "prefecture": "Kyoto"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 192)May include surrounding context.

POST /v1/jp/suppliers/resolve — $0.10

bash
curl -X POST https://api.jpverify.com/v1/jp/suppliers/resolve \
  -H "Authorization: Bearer $JPVERIFY_API_KEY" \
  -H "X-JPVerify-Client: jpverify/1.3.5" \
  -H "Content-Type: application/json" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 203)May include surrounding context.

md
import os, requests

response = requests.post(
    "https://api.jpverify.com/v1/jp/suppliers/resolve",
    headers={"Authorization": f"Bearer {os.environ['JPVERIFY_API_KEY']}",
             "X-JPVerify-Client": "jpverify/1.3.5"},
    json={"name": "株式会社ABC", "registration_number": "T1013301046627", "response_language": "both"},

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 262)May include surrounding context.

POST /v1/jp/addresses/normalize — $0.005

bash
curl -X POST https://api.jpverify.com/v1/jp/addresses/normalize \
  -H "Authorization: Bearer $JPVERIFY_API_KEY" \
  -H "X-JPVerify-Client: jpverify/1.3.5" \
  -H "Content-Type: application/json" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 273)May include surrounding context.

md
import os, requests

response = requests.post(
    "https://api.jpverify.com/v1/jp/addresses/normalize",
    headers={"Authorization": f"Bearer {os.environ['JPVERIFY_API_KEY']}",
             "X-JPVerify-Client": "jpverify/1.3.5"},
    json={"address": "東京都新宿区西新宿2丁目8番1号", "response_language": "both"},

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 285)May include surrounding context.

POST /v1/jp/addresses/parse — $0.002

bash
curl -X POST https://api.jpverify.com/v1/jp/addresses/parse \
  -H "Authorization: Bearer $JPVERIFY_API_KEY" \
  -H "X-JPVerify-Client: jpverify/1.3.5" \
  -H "Content-Type: application/json" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 296)May include surrounding context.

md
import os, requests

response = requests.post(
    "https://api.jpverify.com/v1/jp/addresses/parse",
    headers={"Authorization": f"Bearer {os.environ['JPVERIFY_API_KEY']}",
             "X-JPVerify-Client": "jpverify/1.3.5"},
    json={"address": "東京都千代田区千代田1丁目1番1号", "response_language": "ja"},

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 331)May include surrounding context.

POST /v1/jp/addresses/geocode — $0.005

bash
curl -X POST https://api.jpverify.com/v1/jp/addresses/geocode \
  -H "Authorization: Bearer $JPVERIFY_API_KEY" \
  -H "X-JPVerify-Client: jpverify/1.3.5" \
  -H "Content-Type: application/json" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 342)May include surrounding context.

md
import os, requests

response = requests.post(
    "https://api.jpverify.com/v1/jp/addresses/geocode",
    headers={"Authorization": f"Bearer {os.environ['JPVERIFY_API_KEY']}",
             "X-JPVerify-Client": "jpverify/1.3.5"},
    json={"address": "東京都千代田区千代田1丁目1番1号", "max_candidates": 3},

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 354)May include surrounding context.

POST /v1/jp/addresses/validate — $0.005

bash
curl -X POST https://api.jpverify.com/v1/jp/addresses/validate \
  -H "Authorization: Bearer $JPVERIFY_API_KEY" \
  -H "X-JPVerify-Client: jpverify/1.3.5" \
  -H "Content-Type: application/json" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 365)May include surrounding context.

md
import os, requests

response = requests.post(
    "https://api.jpverify.com/v1/jp/addresses/validate",
    headers={"Authorization": f"Bearer {os.environ['JPVERIFY_API_KEY']}",
             "X-JPVerify-Client": "jpverify/1.3.5"},
    json={"address": "東京都千代田区千代田1丁目1番1号", "required_level": 8, "strict": False},

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 143)May include surrounding context.

API key:

bash
curl -X POST https://api.jpverify.com/v1/jp/companies/resolve \
  -H "Authorization: Bearer $JPVERIFY_API_KEY" \
  -H "X-JPVerify-Client: jpverify/1.3.5" \
  -H "Content-Type: application/json" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
70% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 153)May include surrounding context.

python
import os, requests

response = requests.post(
    "https://api.jpverify.com/v1/jp/companies/resolve",
    headers={"Authorization": f"Bearer {os.environ['JPVERIFY_API_KEY']}",
             "X-JPVerify-Client": "jpverify/1.3.5"},

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 153)May include surrounding context.

python
import os, requests

response = requests.post(
    "https://api.jpverify.com/v1/jp/companies/resolve",
    headers={"Authorization": f"Bearer {os.environ['JPVERIFY_API_KEY']}",
             "X-JPVerify-Client": "jpverify/1.3.5"},

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 202)May include surrounding context.

python
import os, requests

response = requests.post(
    "https://api.jpverify.com/v1/jp/suppliers/resolve",
    headers={"Authorization": f"Bearer {os.environ['JPVERIFY_API_KEY']}",
             "X-JPVerify-Client": "jpverify/1.3.5"},

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/examples.md (reported line 272)May include surrounding context.

python
import os, requests

response = requests.post(
    "https://api.jpverify.com/v1/jp/addresses/normalize",
    headers={"Authorization": f"Bearer {os.environ['JPVERIFY_API_KEY']}",
             "X-JPVerify-Client": "jpverify/1.3.5"},