OpenTangl

WarnAudited by ClawScan on May 10, 2026.

Overview

OpenTangl is openly an autonomous development workflow, but it asks to use authenticated GitHub tooling to create and merge changes across repos without enough visible per-merge guardrails.

Install and run this only on repositories you are willing to let an autonomous tool modify. Use least-privilege GitHub credentials, require branch protection and manual approval before merges, review generated configuration and vision files, and inspect the upstream OpenTangl repository before running npm install.

Findings (5)

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

A bad generated change could be committed or merged into a real repository without the user reviewing that specific change first.

Why it was flagged

The skill is explicitly designed to perform high-impact repository actions automatically; the visible instructions do not show a mandatory human approval step before each merge.

Skill content
it plans features, writes code, verifies builds, creates PRs, reviews diffs, and merges — autonomously
Recommendation

Use branch protections, require manual PR approval before merge, run in a test repository first, and restrict the tool to selected repos and branches.

What this means

If the GitHub session has broad permissions, OpenTangl could act with those permissions across more repositories than the user intended.

Why it was flagged

This shows the workflow relies on the user's authenticated GitHub CLI identity for write/merge actions, but the artifacts do not describe required token scopes, allowed repositories, or privilege limits.

Skill content
GitHub CLI — run `gh auth status` and show the output (needed for PR creation and merging)
Recommendation

Use a dedicated GitHub account or token with least privilege, limit repository access, and confirm the active gh account before running autonomous workflows.

What this means

One bad instruction or generated change could affect multiple repositories and branches, increasing recovery effort.

Why it was flagged

Multi-repository autonomous changes can propagate a mistaken plan, generated bug, or unsafe dependency change across several codebases.

Skill content
Autonomous builds, PRs, reviews, and merges across multiple repos... Manages multiple repos as one product.
Recommendation

Start with one non-production repo, require separate approvals per repository, and keep rollback plans and CI protections in place.

What this means

The reviewed skill file is not the same as reviewing the external OpenTangl code and its dependencies.

Why it was flagged

The skill directs the user to install external project code and npm dependencies manually; this is disclosed and not run automatically by the skill, but that upstream code was not included in the reviewed artifacts.

Skill content
git clone https://github.com/8co/opentangl.git
cd opentangl
npm install
Recommendation

Inspect the upstream repository, package files, and lockfile before installing, and install in an isolated environment if possible.

What this means

Incorrect or manipulated project priorities could steer future generated changes in unwanted directions.

Why it was flagged

OpenTangl maintains persistent project context that can influence later autonomous development runs.

Skill content
Current Priorities (maintained by OpenTangl after each run)
Recommendation

Review the product-vision and project configuration files before each run, especially after automated updates.