Openclaw Skill Clawban

PassAudited by ClawScan on May 10, 2026.

Overview

This appears to be a disclosed Kanban/project-management automation skill, but it can read and change tickets using your existing PM credentials and can optionally run scheduled automation.

Install only if you want an agent to interact with your project-management system. Before use, verify the selected adapter scope, use least-privilege credentials, keep logs protected, and enable the optional cron/autopilot behavior only if you want ongoing scheduled automation.

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

The skill can make visible changes in your project-management tools if invoked with write-capable credentials.

Why it was flagged

The skill intentionally exposes PM workflow actions through external CLIs, including actions that can post comments, move tasks, and create tasks. This is purpose-aligned but user-impacting.

Skill content
It provides: ... Adapters that call existing CLIs using the user’s authenticated session ... start/update/ask/complete/create
Recommendation

Use least-privilege CLI sessions or tokens and review commands before allowing automated write actions.

What this means

If your CLI account has broad access, the skill may be able to read or change more tickets/projects than intended.

Why it was flagged

The skill does not create its own permission boundary; it inherits authority from authenticated PM CLIs or API keys.

Skill content
Privilege inheritance: the skill can do anything the authenticated CLI can do. Example: if `gh` can edit issues and projects in a repo, this skill can too.
Recommendation

Authenticate adapter CLIs with scoped accounts/tokens limited to the intended repo, board, project, workspace, or team.

What this means

Users may not see all required local tools or credentials from registry metadata alone.

Why it was flagged

The registry metadata under-declares runtime/install requirements compared with SKILL.md and package.json, which describe Node/npm, npm ci, external adapter CLIs, and optional API-key env vars.

Skill content
Install specifications: No install spec — this is an instruction-only skill. Code file presence: 38 code file(s). Requirements: Required binaries ... none; Env var declarations: none.
Recommendation

Review SKILL.md and README setup requirements before installing; maintainers should align registry metadata with the documented requirements.

What this means

Sensitive PM content may appear in agent context, stdout/stderr, or logs, and ticket text may influence the agent’s behavior.

Why it was flagged

The skill is designed to bring ticket bodies and comments, including potentially private/internal comments, into agent-visible output.

Skill content
`show` ... Output: title, current stage, URL, full body/description ... and the last 10 comments ... including private/internal comments where supported.
Recommendation

Treat ticket bodies/comments as untrusted and sensitive; avoid running this on boards with secrets unless logs and agent context are appropriately protected.

What this means

Autopilot may continue polling or making workflow-related changes on a schedule until the cron job is removed or disabled.

Why it was flagged

The skill can optionally create scheduled automation that continues running after setup.

Skill content
`--autopilot-install-cron` (creates an OpenClaw cron job that runs `kanban-workflow autopilot-tick`)
Recommendation

Only enable the cron option if you want ongoing automation, and document how to disable or audit the scheduled job.