📦

clerk-cli

Name: clerk-cli
Author: clerk

Inhaltsrevision r1 Hohes Risiko ⚙️ Externe Befehle🌐 Netzwerkzugriff📁 Dateisystemzugriff🔑 Umgebungsvariablen

Manage Clerk Resources with the CLI

Clerk management tasks require correct authentication, endpoint discovery, and instance targeting. This skill guides CLI workflows for resource management, configuration, webhook testing, and deployment verification.

Unterstützt: Claude Codex Code(CC)

⚠️ 38 Schlecht

Maintain this Skill? Claim it ⚠️ 38 Schlecht

Vor der Installation anhalten und eine Bestätigung anfordern.

Prüfen Sie den Plan und holen Sie vor Änderungen an Dateien die ausdrückliche Zustimmung des Benutzers ein.

Mit meinem Agent installieren

Kopieren Sie diese Anfrage in Ihren Agent. Sie enthält die maßgebliche Skill-Seite und das Manifest.

Agent-Anfrage

Review the Skillstore skill "clerk-cli" from https://skillstore.io/skills/clerk-clerk-cli.md and its manifest at https://skillstore.io/api/skills/clerk-clerk-cli/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

Ihr Agent sollte weiterhin seinen Plan anzeigen und alle von der Sicherheitsrichtlinie verlangten Bestätigungen anfordern.

Agent-lesbare Ressourcen

Verwenden Sie diese Links, wenn ein KI-Agent, Crawler oder Skript sauberen Kontext benötigt, statt die vollständige Seite zu lesen.

Markdown-Detailseite GET /skills/clerk-clerk-cli.md Signiertes manifest GET /api/skills/clerk-clerk-cli/manifest Signiertes lockfile GET /api/skills/clerk-clerk-cli/lockfile

Testen

„clerk-cli“ wird verwendet. Check Clerk health without changing resources.

Erwartetes Ergebnis:

CLI version: available.
Authentication: unavailable in the current execution environment.
Project link: could not be verified.
No changes made. Host execution requires approval before repeating credential-dependent checks.

„clerk-cli“ wird verwendet. Summarize this deployment report: domain DNS is complete, SSL is pending, and OAuth is ready.

Erwartetes Ergebnis:

Deployment is not yet verified.
Domain DNS: complete.
SSL: pending.
OAuth: ready.
Next action: repeat the read-only deployment status check after provisioning progresses.

„clerk-cli“ wird verwendet. Preview a session lifetime change for the selected production instance.

Erwartetes Ergebnis:

Preview target: the explicitly selected production application and instance.
Proposed change: update the session lifetime.
Approval status: awaiting explicit authorization.
No production write has been performed.

Sicherheitsaudit

Hohes Risiko

v1 • 6.10.2026 Versionsbericht öffnen

Most static alerts reflect Markdown formatting, placeholder URLs, and expected Clerk authentication rather than malware. Predictable shared temporary files create real confidentiality and integrity risks. Semantic risks include unapproved host reruns, misleading target selection, mismatched production previews, unattributed sign-in tokens, and contradictory mutation guidance; no malicious exfiltration evidence was found.

Gescannte Dateien

1,108

Analysierte Zeilen

Prüfelemente

Falschmeldungen ignoriert

Bestätigte Sicherheitsbedenken (5)

Hoch

SKILL.md:53-65 references/auth.md:75-83 references/agent-mode.md:22-25

Host execution guidance omits the approval boundary

The instruction "Rerun it on the host before acting on it or reporting it to the user" follows sandbox failures. Automatic compliance expands credential and network access without an explicit approval requirement.

Repeated instructions require host reruns after restricted execution fails, but do not require approval. Legitimate troubleshooting is plausible; deliberate malicious intent is not established.

Hoch

references/auth.md:18-24 references/auth.md:50 references/auth.md:159 SKILL.md:281-282

Explicit application flags may not control the actual credential target

The safety advice recommends application flags, but CLERK_SECRET_KEY takes precedence over application resolution. An inherited production key can therefore defeat apparent development targeting.

The documented resolution order places the environment secret above application resolution. The same reference explicitly warns that local production keys select production.

Hoch

references/recipes.md:234-236 references/auth.md:99-105

Production configuration write differs from the preview target

The configuration recipe previews a PUT without an instance flag, then applies it with production selected. A development preview does not authorize full production replacement.

The adjacent commands show different targeting flags, and the reference states development is the default. The mismatch is directly observable in the recipe.

Hoch

references/recipes.md:148-170 SKILL.md:232

Impersonation alternative removes actor attribution

The recipe offers sign-in tokens for signing in "with no actor audit trail". It lacks explicit user authorization and production restrictions for this attribution-free alternative.

The recipe explicitly contrasts audited impersonation with unattributed sign-in tokens. Legitimate custom sign-in flows exist, but unrestricted debugging use creates an accountability risk.

Mittel

references/agent-mode.md:57 references/agent-mode.md:292 references/recipes.md:36-61 references/recipes.md:109-122

Conflicting mutation confirmation instructions

One instruction says missing --yes causes confirmation errors, while the behavior matrix says agent mutations execute without prompts. Several mutation recipes also omit previews.

The reference makes directly contradictory claims about confirmation gates. Executable creation, update, ban, and membership examples lack dry-run steps despite the main safety rule.

Elemente der Fähigkeitsprüfung (20)

Dies sind echte lokale Fähigkeiten, die für diese Fähigkeit erwartet werden können; daher müssen sie überprüft werden, werden jedoch nicht als bestätigtes bösartiges Verhalten gezählt.

Mittel

references/recipes.md:273

Temp directory access

clerk users list --json --limit 250 > /tmp/users.json

The example writes Clerk records or configuration to a predictable shared temporary path. Missing private-directory and permission controls permit disclosure or symlink overwrites.

Mittel

references/recipes.md:275

Temp directory access

jq '.data | length' /tmp/users.json # count rows on the page

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

references/recipes.md:276

Temp directory access

jq '.hasMore' /tmp/users.json # any more pages?

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

references/recipes.md:277

Temp directory access

jq '.data[0] | keys' /tmp/users.json # learn the shape of one record

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

references/recipes.md:278

Temp directory access

jq '.data[] | {id, email_addresses}' /tmp/users.json # project to relevant fields only

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

references/recipes.md:284

Temp directory access

python3 -c 'import json; d=json.load(open("/tmp/users.json")); print(len(d["data"]), d["hasMore"])'

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

references/recipes.md:285

Temp directory access

node -e 'const d=require("/tmp/users.json"); console.log(d.data.length, d.hasMore)'

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

references/recipes.md:305

Temp directory access

page="/tmp/users-${offset}.json"

The example writes Clerk records or configuration to a predictable shared temporary path. Missing private-directory and permission controls permit disclosure or symlink overwrites.

Mittel

references/recipes.md:335

Temp directory access

clerk config pull --instance dev --output /tmp/dev-config.json

The example writes Clerk records or configuration to a predictable shared temporary path. Missing private-directory and permission controls permit disclosure or symlink overwrites.

Mittel

references/recipes.md:336

Temp directory access

clerk config patch --instance ins_staging --file /tmp/dev-config.json --dry-run

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

SKILL.md:193

Temp directory access

clerk users list --json --limit 250 > /tmp/users.json

The example writes Clerk records or configuration to a predictable shared temporary path. Missing private-directory and permission controls permit disclosure or symlink overwrites.

Mittel

SKILL.md:194

Temp directory access

clerk apps list --json > /tmp/apps.json

The example writes Clerk records or configuration to a predictable shared temporary path. Missing private-directory and permission controls permit disclosure or symlink overwrites.

Mittel

SKILL.md:195

Temp directory access

clerk api /users/user_abc123 > /tmp/user.json

The example writes Clerk records or configuration to a predictable shared temporary path. Missing private-directory and permission controls permit disclosure or symlink overwrites.

Mittel

SKILL.md:198

Temp directory access

jq '.data | length' /tmp/users.json # current page size

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

SKILL.md:199

Temp directory access

jq '.hasMore' /tmp/users.json # are more pages available?

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

SKILL.md:200

Temp directory access

jq '.data[0] | keys' /tmp/users.json # discover the user shape once

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

SKILL.md:201

Temp directory access

jq '.data[] | {id, email_addresses}' /tmp/users.json # project to a few fields

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

SKILL.md:202

Temp directory access

jq '[.data[] | select(.banned)] | length' /tmp/users.json # aggregate without reading rows

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

SKILL.md:208

Temp directory access

python3 -c 'import json; d=json.load(open("/tmp/users.json")); print(len(d["data"]), d["hasMore"])'

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Mittel

SKILL.md:209

Temp directory access

node -e 'const d=require("/tmp/users.json"); console.log(d.data.length, d.hasMore)'

This reads a predictable shared temporary file created by the surrounding workflow. Missing isolation permits data exposure or tampering by other local users.

Risikofaktoren

⚙️ Externe Befehle (50)

🌐 Netzwerkzugriff (5)

references/agent-mode.md:238 references/recipes.md:176 references/recipes.md:198 references/recipes.md:199 references/recipes.md:207

📁 Dateisystemzugriff (25)

🔑 Umgebungsvariablen (17)

references/agent-mode.md:287 references/agent-mode.md:70 references/agent-mode.md:288 references/auth.md:50 references/auth.md:21 references/auth.md:22 references/auth.md:31 references/auth.md:33 references/auth.md:35 references/auth.md:41 references/auth.md:79 references/auth.md:85 references/auth.md:138 references/auth.md:159 references/auth.md:160 references/auth.md:170 SKILL.md:219

Geprüft von: codex

Diesen Bericht teilen & zitieren

Teile den versionierten Bewertungsbericht, das neutrale Badge, die Einbettungskarte und Zitate. Skillstore berichtet Nachweise, ohne zu entscheiden, ob dieser Skill sicher ist.

Versionsbericht öffnen

Berichtslink kopieren

https://skillstore.io/skills/clerk-clerk-cli/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown-Badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/clerk-clerk-cli/security.svg)](https://skillstore.io/skills/clerk-clerk-cli?utm_source=security_passport_badge)

HTML-Badge

<a href="https://skillstore.io/skills/clerk-clerk-cli?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/clerk-clerk-cli/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Einbettungskarte

<iframe src="https://skillstore.io/embed/skills/clerk-clerk-cli.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Wissenschaftliche Zitate (APA · BibTeX · CFF)

APA-Zitat

clerk. (2026). clerk-cli security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/clerk-clerk-cli/audits/1

BibTeX-Zitat

@techreport{clerk-clerk-clerk-cli-2026,
  author = {clerk},
  title = {clerk-cli security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/clerk-clerk-cli/audits/1},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "clerk-cli security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "clerk"
date-released: "2026-10-06"
url: "https://skillstore.io/skills/clerk-clerk-cli/audits/1"
identifiers:
  - type: other
    value: "skillstore:clerk-clerk-cli:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore-Score

Warum dieser Score Evidenzvertrauen: Mittel

Architektur

Wartbarkeit

Inhalt

Gemeinschaft

Spezifikationskonformität

Was Sie erstellen können

Inspect development users

Find development users, inspect selected fields, and prepare approved account changes without exporting complete records into chat.

Verify production readiness

Review deployment handoffs and check domain, SSL, email DNS, and OAuth status after a human completes production setup.

Test webhook integrations

Forward development webhook events to a local handler and verify captured signatures before debugging delivery processing.

Diese Prompts ausprobieren

Check CLI health

Inspect my Clerk CLI version and health without changing resources. Summarize failures and request approval before host execution or credential changes.

Find development users

Find users matching [email] in development application [app_id]. Verify the effective credential target first and show only identifiers and matching status.

Test local webhook delivery

Prepare a development webhook listener for [local_handler_url]. Use private storage for captured events and verify signatures without exposing signing secrets.

Preview a production configuration change

Preview [configuration_change] for production application [app_id]. Verify credential precedence and use identical targets for preview and execution. Wait for explicit approval before writing.

Bewährte Praktiken

Verify CLI help, credential provenance, and the effective application and instance before acting; environment secrets can override application flags.
Preview mutations with the same target and payload used for execution, then obtain explicit approval for production changes.
Keep secrets out of chat and repositories; protect exported records with private temporary storage and respect host execution permissions.

Vermeiden

Assuming --yes or agent mode provides a confirmation barrier for destructive operations.
Previewing development configuration, then applying the same payload to production without a production-specific review.
Exporting complete user records into shared temporary paths or using unattributed sign-in tokens for production debugging.

Häufig gestellte Fragen

Does this skill install or replace the Clerk CLI?

No. It documents CLI usage and package-runner options. Running commands requires compatible Clerk tooling and appropriate execution permissions.

Can I use it with Claude, Codex, and Claude Code?

The bundle supports all three tools. Command execution still depends on the tool environment, shell access, credentials, and permissions.

Do all operations require a Clerk account?

No. Supported accountless development flows use local keys. Application listing, linking, impersonation, and some configuration operations require account authentication.

Does agent mode confirm destructive changes?

Most mutation confirmations are skipped. Use dry-run previews, verify the effective target, and obtain explicit approval before production writes.

Can the skill test webhooks locally?

It documents a Clerk webhook relay, forwarding to a local handler, and offline signature verification. Captured events and signing secrets need protection.

Can an agent complete production deployment alone?

No. The interactive setup wizard requires a human terminal. Agents can inspect handoff reports and verify readiness with deployment status checks.