Coder for OpenClaw
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This is a transparent installer for a powerful coding sub-agent, but it relies on an external unpinned repository, Docker, and an authenticated provider profile that users should review before use.
Before installing, inspect the linked GitHub repository, Dockerfile, agent config, and runtime prompt; ideally pin a known-good commit. Use a dedicated provider profile, keep work in the sandbox, and review the persistent OpenClaw agent registration.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The code and Dockerfile that actually define the runtime could change independently of this skill package.
The reviewed package does not include the repository contents; the user is instructed to clone an external repo at its current state and build a Docker image from it.
`git clone https://github.com/MilleniumGenAI/coder-openclaw-agent.git` ... `docker build -f docker/coder-sandbox.dockerfile -t coder-sandbox:latest .`
Inspect the repository before use and consider pinning to a trusted commit or release instead of building from an unpinned branch.
A coding agent can run commands and modify generated or mounted work products if tasks are delegated to it.
The skill intentionally wires a sub-agent that can execute code. The artifact presents this as sandboxed and purpose-aligned, but it is still a high-impact capability.
- code execution and verification inside the OpenClaw sandbox;
Use it only in a sandboxed workspace, keep task scopes narrow, and review outputs before applying changes to important projects.
Delegated coding tasks may consume provider quota and send task content to the configured provider.
The skill requires an authenticated provider profile, which is expected for a model-backed coding agent but gives the agent access to that configured provider account.
- an authenticated `openai-codex` provider profile
Use a dedicated, least-privilege provider profile where possible and avoid delegating sensitive files unless you are comfortable with the provider data flow.
The coder sub-agent may remain available for future delegation until the configuration is removed or disabled.
The skill installs a persistent sub-agent configuration. This is disclosed and expected, but it remains part of OpenClaw's agent setup after installation.
Register the agent in `openclaw.json` using: `openclaw/agent-config.template.json`
Review the generated `openclaw.json` entry and remove or disable the agent when it is no longer needed.