primer-x402
Verdict safe, but 2 findings (2 medium), mainly data exfiltration. Most recent scan 2026-02-07.
The primer-x402 skill is a legitimate cryptocurrency payment integration tool with MODERATE security concerns. The skill enables HTTP-native crypto payments using the x402 protocol and USDC on EVM chains. Primary concerns include: (1) private key handling via environment variables without adequate security warnings, (2) missing allowed-tools declaration despite requiring Bash/Python execution, and (3) incomplete documentation with missing example files. The skill does NOT exhibit malicious behavior such as data exfiltration, prompt injection, or command injection. The core functionality (making crypto payments via CLI tools) is transparent and matches the stated description. The skill appropriately uses external package managers (npx, pip) to install the x402 CLI tool rather than bundling potentially malicious code. However, the financial nature of the skill (handling private keys and payments) means security documentation should be more comprehensive. Recommended actions: add allowed-tools declaration, enhance private key security guidance, complete code examples, and include missing reference files.
Scanner's overall assessment (LLM analyzer)
Second scanner: NVIDIA SkillSpector
Independent static run, no language model, over the same files as the Cisco scan above. How the two scanners compare across the whole corpus: Two scanners, one corpus.
What this skill does
Make HTTP-native crypto payments using the x402 protocol. Pay for APIs, access paid resources, and handle 402 Payment Required responses with USDC on Base and other EVM chains.
- Package
- 2 files, 6.9 KB
- SKILL.md
- 896 words, 12+ sections
- Code
- None. Instructions only.
- Requires
- anyBins:
nodenpxpython3pip
SKILL.md covers: x402 Payment Protocol · When to Use This Skill · How to Respond · Quick Setup · Node.js · Python · How x402 Works · CLI Commands · Examples · Check if a URL requires payment · Preview payment cost (dry run - no payment made) · Check wallet balance
Findings
MEDIUM Missing allowed-tools Declaration Data Exfiltrationllm
The skill does not declare allowed-tools in its YAML manifest. While this field is optional, the skill's instructions clearly require Bash and Python execution (npx, pip, x402 CLI commands). Without explicit tool declarations, users cannot verify the skill's intended tool usage scope.
SKILL.md
allowed-tools: Not specified
But instructions require:
- npx @primersystems/x402 openclaw init
- pip install primer-x402
- x402 wallet create
- Bash command executionMEDIUM Private Key Handling in Environment Variables Data Exfiltrationllm
The skill instructs users to pass private keys via environment variables (X402_PRIVATE_KEY) in command-line examples. While this is a common pattern, it poses risks: environment variables can leak through process listings, shell history, logs, and error messages. The skill does not provide warnings about secure key management or alternative methods.
SKILL.md
# Pay for an API (max $0.10)
X402_PRIVATE_KEY=0x... npx @primersystems/x402 pay https://api.example.com/paid --max-amount 0.10What is in the package
| File | Kind | Size |
|---|---|---|
SKILL.md | doc | 6.5 KB |
_meta.json | config | 405 B |
External hosts referenced
chromewebstore.google.com github.com primer.systems npmjs.com pypi.org x402.org
Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.
Environment variables referenced
X402_PRIVATE_KEY
Package contents recorded 2026-10-22. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.
What we checked
- Static analyzer. Pattern rules over SKILL.md and bundled files: known-bad commands, exfiltration URLs, secrets, obfuscation.
- Llm analyzer. A language model reads the skill and reports intent-level issues the rules cannot express.
- Meta analyzer. A second model pass that validates, deduplicates and re-scores the other analyzers' findings.
What we did not check
- Runtime behaviour. The skill was not executed; findings come from reading its files.
- Anything fetched at install or run time, including remote scripts, package dependencies, and URLs the skill tells the agent to open.
- Content the analyzers cannot read: images, binaries, and some non-English text.
- Known scanner bypass classes. See the methodology and limitations page.
Sources
- Repository: https://clawhub.ai/skills/primer-x402
- Raw scan result: primer-x402-scan.json
- Badge for your README:
- Request a re-scan