Jenkins

WarnAudited by ClawScan on May 10, 2026.

Overview

The skill is coherent for Jenkins access, but it can use a Jenkins API token to start or stop CI/CD jobs without built-in scoping or confirmation.

Review before installing. If you use it, configure a dedicated least-privilege Jenkins token over HTTPS, require explicit confirmation before build/stop actions, limit which jobs or environments may be controlled, and avoid pulling full console logs unless needed.

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

An agent using a sufficiently privileged Jenkins token could start or interrupt CI/CD jobs, which may deploy code or disrupt pipelines.

Why it was flagged

The script directly sends POST requests that trigger and stop Jenkins builds, and the artifacts do not show a confirmation step, allowlist, or environment boundary for these high-impact operations.

Skill content
await request(buildUrl, { method: "POST" }); ... await request(`/job/${encodeURIComponent(job)}/${buildNum}/stop`, { method: "POST" });
Recommendation

Require explicit user approval for build and stop actions, restrict the Jenkins token to least privilege, and consider allowlisting approved jobs or environments.

What this means

The skill can act with the Jenkins permissions of the configured user/token.

Why it was flagged

The script uses a Jenkins API token for Basic Auth. This is purpose-aligned, but the token may carry broad Jenkins permissions and should be scoped carefully.

Skill content
const JENKINS_API_TOKEN = process.env.JENKINS_API_TOKEN; ... "Authorization": `Basic ${auth}`
Recommendation

Use a dedicated least-privilege Jenkins token, prefer HTTPS Jenkins URLs, and avoid using a highly privileged administrator credential.

What this means

Sensitive CI/CD log data could be exposed in the chat or reused by the agent as context.

Why it was flagged

The console command returns Jenkins console output into the agent-visible result. Build logs may contain secrets, internal details, or untrusted text.

Skill content
console.log(JSON.stringify({ job: job, build: last ? "last" : buildNum, console: output }, null, 2));
Recommendation

Use the --tail option when possible, avoid retrieving logs that may contain secrets, and treat console output as untrusted text.

What this means

Users have less external assurance about the maintainer or upstream source of a tool that can affect Jenkins pipelines.

Why it was flagged

For a CI/CD automation skill, limited source provenance reduces assurance even though the included script is visible and no remote installer is shown.

Skill content
Source: unknown; Homepage: none
Recommendation

Review the bundled script before use and prefer a skill with a verifiable source or trusted publisher for production Jenkins access.