Openpond Cli

ReviewAudited by ClawScan on May 10, 2026.

Overview

The skill is a coherent OpenPond CLI guide, but it involves installing an external npm CLI and using OpenPond credentials for commands that can push code, run tools, and create agents.

Before installing, verify the npm package source, use a least-privilege OpenPond API key, review files and diffs before any repo push, and require explicit approval before running tools or creating OpenPond agents.

Findings (4)

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.

What this means

Installing or running the npm package will execute code from outside this skill review.

Why it was flagged

The skill asks the user to install or run an external npm package that is not included in the reviewed artifacts and is not pinned to a version.

Skill content
Install: `npm i -g openpond-code` (or `npx --package openpond-code openpond <cmd>`)
Recommendation

Verify that `openpond-code` is the expected OpenPond CLI package, prefer trusted sources, and pin or review the package version when possible.

What this means

Anyone or anything with access to the configured credential may be able to act on the OpenPond account within that credential's scope.

Why it was flagged

The workflow uses OpenPond account credentials and a local cache, which is expected for CLI account operations but sensitive.

Skill content
Auth: run `openpond login` or set `OPENPOND_API_KEY` ... Cache file: `~/.openpond/cache.json`
Recommendation

Use least-privilege API keys, avoid pasting real keys into shared chats or logs, and review local credential/cache handling.

What this means

An agent running these commands could push unintended files or trigger tool actions if the target path, repo, or tool is wrong.

Why it was flagged

The documented CLI workflows can upload repository contents and run OpenPond tools; these are purpose-aligned but can have account or data side effects.

Skill content
`git add . && git commit -m "init"` ... `openpond repo push --path . --branch main` ... `openpond tool run handle/repo myTool --body ...`
Recommendation

Require explicit user approval for pushes and tool runs, inspect the git diff before pushing, and avoid broad `git add .` when secrets or unrelated files may be present.

What this means

A created agent could continue to exist or act according to its OpenPond configuration after the current task.

Why it was flagged

The skill documents creating an OpenPond agent, which may persist beyond the immediate CLI session.

Skill content
`openpond apps agent create --prompt "Build a daily digest agent"`
Recommendation

Only create agents on explicit request, confirm the prompt and intended lifecycle, and know how to disable or delete created agents.