Contract Interact Checklist for Openclaw

Contract Interact Checklist applies a repeatable preflight risk review to smart contract interactions and fails the check when required safeguards are incomplete.

zhaoxinghua09-cell
v1.0.3
Sep 29, 2026
0
330
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install contract-interact-checklist

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install contract-interact-checklist using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Contract Interact Checklist?

Contract Interact Checklist is a command-line and AI-agent workflow for checking risks before interacting with a smart contract. It covers six areas: contract audit status, address verification, authorization allowance, revocation, private-key safety, and a small test transaction. Its deterministic PASS/FAIL result helps teams apply a consistent safety gate rather than relying on an informal review. Among Openclaw Skills, it is a focused option for adding structured checks to contract-interaction workflows.

The skill is zero-dependency and designed to run locally and offline. It does not collect or upload business data, and it does not store sensitive plaintext, passwords, or private keys. It is an assistant framework, not legal, investment, or professional advice; people must review consequential decisions and verify current official sources.

Contract Interact Checklist Use Cases

  • Run a preflight review before calling or transacting with a smart contract.
  • Check that the intended contract address and audit information have been reviewed.
  • Verify authorization allowances and consider revocation where appropriate.
  • Confirm private keys are not exposed or pasted into webpages or unsafe workflows.
  • Require a small test transaction before proceeding with a larger interaction.
  • Integrate a consistent PASS/FAIL safety gate into an AI-agent or command-line workflow.

How Contract Interact Checklist Works

  1. Invoke the checklist script with the required arguments; use --help to inspect the available options.
  2. Review the interaction against the six safeguards: audit, address, allowance, revocation, private-key handling, and a small test transaction.
  3. Provide the requested information and evidence for the checks. Keep sensitive plaintext and keys out of the workflow; evidence should follow the designated evidence-vault process.
  4. The script returns a deterministic result: rc=0 for PASS, rc=1 when a risk or incomplete check is found, and rc=2 for usage or environment errors.
  5. Optionally request machine-readable output with --json for downstream tooling or agent workflows.
  6. Have a human or qualified professional review important decisions; the checklist does not replace expert judgment or official compliance guidance.

Contract Interact Checklist Setup

Install the skill with the skills CLI or copy it into the skills directory used by your agent. The package supports Windows, macOS, and Linux. Python is needed to run the included script; the skill is described as having zero external dependencies.

Using the skills CLI:

npx [email protected] add https://github.com/zhaoxinghua09-cell/agent-skills/tree/v2026.09.28.1/skills/contract-interact-checklist

Or clone the published revision and copy the skill into an agent skills directory:

git clone --branch v2026.09.28.1 --depth 1 https://github.com/zhaoxinghua09-cell/agent-skills.git
git -C agent-skills rev-parse HEAD
cp -r agent-skills/skills/contract-interact-checklist ~/.workbuddy/skills/

For the manual install, verify that the cloned commit matches the published tag v2026.09.28.1. For Claude Code, the documented location is ~/.claude/skills/contract-interact-checklist/; Codex, Cursor, and WorkBuddy use their respective skills directories. Then inspect the script options and run the checklist:

python scripts/contract_interact_checklist.py --help
python scripts/contract_interact_checklist.py --json

The second command illustrates the JSON-output flag; provide the required arguments shown by --help for an actual check. Exit codes are 0 for PASS, 1 for a detected issue, and 2 for usage or environment errors.

Contract Interact Checklist Data Schema & Taxonomy

The skill defines a checklist workflow and machine-readable output option, but the supplied documentation does not specify a persistent record format, generated report files, or a formal metadata schema.

Element Documented behavior
Checklist dimensions Audit, contract address, authorization allowance, revocation, private-key safety, and small test transaction
Output Deterministic result; optional machine-readable output via --json
Exit status rc=0 PASS; rc=1 risk found or check failed; rc=2 usage or environment error
Sensitive information Do not store sensitive plaintext, passwords, or private keys; evidence is directed to an evidence vault
Data handling Intended for local offline operation; documentation says user business data is not collected or uploaded
Files and persistence No additional generated files or persistence behavior are specified in the provided SKILL.md

The Python code file is licensed MIT. The skill documentation and theory text are explicitly outside that code license, so review the package's license terms before redistributing documentation.

Contract Interact Checklist Advanced Features

  • Deterministic exit codes make the checklist suitable for shell scripts and automated gates: PASS (0), failed checks (1), and usage or environment errors (2).
  • Optional --json output supports machine-readable handling by compatible agent or command-line workflows.
  • Zero external dependencies and local, offline operation reduce setup overhead and avoid uploading business data, according to the documentation.
  • Human-readable error messages are used for failures without exposing raw stack traces.
  • The skill can be placed in the skills directories used by Claude Code, Codex, Cursor, or WorkBuddy; the documented workflow treats SKILL.md as the agent procedure.
  • It is a compliance pre-screening aid, not an autonomous approval system: key judgments require human review, and current official sources and qualified professionals take precedence.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*