Skills nla-create
📦

nla-create

v1.0 Content revision r1 High Risk ⚙️ External commands🔑 Env variables

Create NLA Blockchain Escrows

Creating a token escrow with natural language conditions requires exact parameters and careful wallet setup. This skill guides Claude, Codex, or Claude Code through requirement gathering, demand drafting, checks, and nla CLI execution.

Supports: Claude Codex Code(CC)
⚠️ 38 Poor

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request
Review the Skillstore skill "nla-create" from https://skillstore.io/skills/internet-court-nla-create.md and its manifest at https://skillstore.io/api/skills/internet-court-nla-create/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/internet-court-nla-create.md Signed manifest GET /api/skills/internet-court-nla-create/manifest Signed lockfile GET /api/skills/internet-court-nla-create/lockfile

Test it

Using "nla-create". I want to escrow tokens for someone who submits a useful technical report.

Expected outcome:

The skill asks for the token, amount, oracle, network, and report acceptance criteria, then rewrites the demand for clearer arbitration.

Using "nla-create". Create a Sepolia escrow using Anthropic arbitration.

Expected outcome:

The skill checks the active network and wallet, confirms the provider and model, then prepares the escrow creation command for review.

Using "nla-create". The command returned an escrow UID.

Expected outcome:

The skill highlights the UID, explains why it must be saved, and describes how fulfillment and collection use that identifier.

Security Audit

High Risk
v2 • 7/21/2026 Open versioned report

The skill contains legitimate documentation and commands for an external blockchain CLI, but several Markdown backtick detections are false positives. Confirmed findings concern execution of CLI commands, wallet private-key handling, and escrow creation that can lock ERC20 tokens on-chain. The workflow should require explicit confirmation before any state-changing or mainnet action.

1
Files scanned
104
Lines analyzed
11
Review items
0
False positives ignored

Confirmed security concerns (1)

High
SKILL.md:62-73
Irreversible Asset-Locking Transaction Without Confirmation Step
The workflow directs execution of nla escrow:create, which can lock ERC20 tokens on-chain, but it does not require a final user confirmation that shows the network, token, amount, and recipient oracle.
The documented command creates an escrow using user-provided asset and network parameters, and the workflow moves directly from gathering requirements to execution without a final confirmation gate.
Capability review items (11)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

High
SKILL.md:52
Generic API/secret keys
- Set the `PRIVATE_KEY` environment variable
The skill explicitly asks users to set PRIVATE_KEY for CLI consumption. A wallet private key is a high-value credential whose exposure permits asset control.
Medium
SKILL.md:38-47
Ruby/shell backtick execution
```bash
This begins a Bash block that directs the agent to run nla commands. Although the listed checks are low impact, external CLI execution can act on the local environment.
Medium
SKILL.md:47-50
Ruby/shell backtick execution
```
This closes a Bash block containing external nla commands. The skill authorizes command execution rather than only providing conceptual guidance.
Medium
SKILL.md:50-51
Ruby/shell backtick execution
- Run `nla wallet:set --private-key <key>`
The instruction places a private key directly in a CLI command. Command-line secrets can be exposed through shell history and process listings.
Medium
SKILL.md:51-52
Ruby/shell backtick execution
- Pass `--private-key <key>` to the command
Passing a private key as a command argument creates the same command-line secret exposure risk. The command may also configure a wallet for later transactions.
Medium
SKILL.md:52-64
Ruby/shell backtick execution
- Set the `PRIVATE_KEY` environment variable
The skill instructs users to provide a private key through an environment variable for CLI use. This is sensitive credential handling tied to financial transactions.
Medium
SKILL.md:64-73
Ruby/shell backtick execution
```bash
This begins the nla escrow:create command block. It initiates an on-chain escrow creation flow that can lock user-controlled ERC20 assets.
Medium
SKILL.md:73-77
Ruby/shell backtick execution
```
This closes the escrow creation command block, which includes user-supplied transaction parameters. Executing it can cause an on-chain financial commitment.
Medium
SKILL.md:81
Ruby/shell backtick execution
- Available networks: `anvil` (local), `sepolia`, `base-sepolia`, `mainnet`. Switch with `nla switch
The line instructs users to switch the nla CLI network, including mainnet. Network selection changes where subsequent financial commands execute.
Medium
SKILL.md:82-83
Ruby/shell backtick execution
- For local dev, `nla dev` starts Anvil, deploys contracts, creates mock tokens, and starts the orac
The nla dev command starts local services and deploys contracts. It is an external command that changes the development environment.
Medium
SKILL.md:95-103
Ruby/shell backtick execution
```bash
This begins an example Bash command that creates an escrow with specified token and oracle addresses. Running it can submit a blockchain transaction.

Risk Factors

⚙️ External commands (19)
SKILL.md:13 SKILL.md:28 SKILL.md:32 SKILL.md:33 SKILL.md:34 SKILL.md:38-47 SKILL.md:47-50 SKILL.md:50-51 SKILL.md:51-52 SKILL.md:52-64 SKILL.md:64-73 SKILL.md:73-77 SKILL.md:77-81 SKILL.md:81 SKILL.md:82-83 SKILL.md:83-85 SKILL.md:85-91 SKILL.md:91-95 SKILL.md:95-103
🔑 Env variables (1)
SKILL.md:52
Audited by: claude View Audit History →
Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report
Security Assessment

Copy report link

https://skillstore.io/skills/internet-court-nla-create/audits/2?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/internet-court-nla-create/security.svg)](https://skillstore.io/skills/internet-court-nla-create?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/internet-court-nla-create?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/internet-court-nla-create/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/internet-court-nla-create.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
Academic citations (APA · BibTeX · CFF)

APA citation

internet-court. (2026). nla-create security audit report (audit version 2) [Author version 1.0]. Skillstore. https://skillstore.io/skills/internet-court-nla-create/audits/2

BibTeX citation

@techreport{internet-court-internet-court-nla-create-2026, author = {internet-court}, title = {nla-create security audit report (audit version 2)}, institution = {Skillstore}, year = {2026}, number = {2}, url = {https://skillstore.io/skills/internet-court-nla-create/audits/2}, note = {Author version 1.0} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "nla-create security audit report (audit version 2)" version: "1.0" type: report authors: - name: "internet-court" date-released: "2026-07-21" url: "https://skillstore.io/skills/internet-court-nla-create/audits/2" identifiers: - type: other value: "skillstore:internet-court-nla-create:audit:2" description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium
55
Architecture
85
Maintainability
87
Content
65
Community
83
Spec Compliance

What You Can Build

Launch a conditional token bounty

Create an escrow that releases ERC20 tokens when a submitted fulfillment satisfies a written demand.

Test NLA workflows on a development network

Use local or testnet settings to verify networks, wallets, contracts, and oracle behavior before mainnet use.

Draft clearer oracle arbitration terms

Turn vague escrow goals into specific demands that are easier for an AI oracle to judge.

Try These Prompts

Create my first escrow 
Help me create an NLA escrow. Ask for each required value, check my setup, and explain every step before running commands.
Draft a stronger demand 
Review this escrow demand for clarity, testability, and public visibility. Suggest a tighter version before I create the escrow.
Prepare a testnet escrow 
Guide me through creating an NLA escrow on a test network. Confirm the network, token, oracle, wallet, and arbitration model first.
Configure advanced arbitration 
Help me create an escrow with a custom arbitration provider, model, and prompt. Validate the placeholders and summarize all public on-chain fields.

Best Practices

  • Use testnets or local development before locking valuable tokens on mainnet.
  • Keep demands specific, verifiable, and free of secrets because terms may be public.
  • Confirm network, token address, oracle address, amount, and wallet before execution.

Avoid

  • Do not paste private keys into chat or store them in shell history.
  • Do not create escrows with vague demands that an oracle cannot evaluate.
  • Do not run mainnet commands until testnet behavior and wallet balances are verified.

Frequently Asked Questions

What is an NLA escrow?
It is an on-chain escrow that locks ERC20 tokens until an AI oracle judges a natural language demand as fulfilled.
Which tools can use this skill?
The report lists support for Claude, Codex, and Claude Code.
Does the skill convert token decimals?
No. It expects the amount in the token smallest unit, so users must calculate the value before execution.
Can escrow terms include private information?
No. Demand text, provider, model, and prompt settings can be encoded on-chain and visible to others.
What networks are documented?
The skill documents anvil, sepolia, base-sepolia, and mainnet, with a command to switch networks.
What should I verify before creating an escrow?
Verify the wallet, network, token address, amount, oracle address, provider, model, prompt, and transaction consequences.

Developer Details

License

MIT

Author version

v1.0

Skillstore revision

r1

Version notice

The author-declared version is not valid SemVer.

Repository

https://github.com/internet-court/internet-court-skill/tree/main/vendored/arkhai/nla-create

Ref

3f6e026a3363e0954ede7bef0cfe88d4475de137

Maintenance freshness

7/21/2026

Usage

1 downloads · 0 views

File structure

📄 SKILL.md