Back to skill

Security audit

design-ads 海报制作

Security checks for vulnerabilities and agentic risk

Overview

The skill has a coherent poster-generation purpose, but its renderer runs active HTML in an unsandboxed browser and has avoidable dependency and file-copying risks that warrant review before installation.

Review this skill before installing, especially if it may render HTML supplied by others or content containing external URLs. Run it only in a low-privilege, isolated environment, avoid sensitive data in poster HTML, and prefer a version that keeps the Chromium sandbox enabled, blocks unexpected network requests, pins dependencies, and stops automatically exporting source HTML unless requested.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Findings (2)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/render.js:125
Finding

Caller-Controlled HTML Is Rendered with the Chromium Sandbox Disabled

Content
View full analysis
Remediation
View remediation
{ const url = new URL(request.url()); if (url.protocol === 'data:' || url.protocol === 'about:') { request.continue(); return; } if (url.protocol === 'file:' && url.pathname === absoluteHtmlPath) { request.continue(); return; } request.abort(); }); ``` 7. If local fonts or assets are needed, copy them into a dedicated temporary directory and allow only canonical paths within that directory. 8. Validate and canonicalize the input path before loading it. Reject symbolic links and paths outside the designated input directory where feasible. 9. Apply a restrictive Content Security Policy to generated HTML, such as `default-src 'none'; style-src 'unsafe-inline'; img-src data: file:; font-src file:`, adjusted only for required resources. 10. Keep Chrome or Chromium fully patched and pin the renderer to a reviewed, supported browser version. ]]>

T08 · Insecure Dependencies

Warning
Location
package-lock.json:816
Finding

Puppeteer Installation Uses a Third-Party Package Mirror and Executes an Install Script

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (19)

Known Vulnerable Dependency: basic-ftp==5.2.0 — 4 advisory(ies): GHSA-6v7q-wjvx-w8wg (basic-ftp: Incomplete CRLF Injection Protection Allows Arbitrary FTP Command Exe); CVE-2026-39983 (basic-ftp has FTP Command Injection via CRLF); CVE-2026-41324 (basic-ftp vulnerable to denial of service via unbounded memory consumption in Cl) +1 more

High
Category
Supply Chain
Confidence
80% confidence
Finding

Dependency has known vulnerabilities (CVEs). Using packages with unpatched security flaws exposes the environment to known exploits.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: extract-zip==2.0.1 — 2 advisory(ies): CVE-2026-19693 (extract-zip allows arbitrary file writes through symlink archive entries); CVE-2026-56876 (extract-zip unvalidated symlink path traversal)

High
Category
Supply Chain
Confidence
88% confidence
Finding

extract-zip is a real transitive dependency in this lockfile and has file-write/path-traversal style advisories involving symlink archive entries. Because Puppeteer commonly downloads and extracts browser archives during install/runtime tooling, a compromised or malicious archive source could potentially lead to arbitrary file writes on the host, though the skill context does not suggest the skill itself handles attacker-supplied zip files directly.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: ip-address==10.1.0 — 2 advisory(ies): CVE-2026-69192 (ip-address: Address4 decodes leading-zero octets as decimal while resolvers deco); CVE-2026-42338 (ip-address has XSS in Address6 HTML-emitting methods)

High
Category
Supply Chain
Confidence
80% confidence
Finding

Dependency has known vulnerabilities (CVEs). Using packages with unpatched security flaws exposes the environment to known exploits.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: js-yaml==4.1.1 — 4 advisory(ies): CVE-2026-84375 (js-yaml: maxTotalMergeKeys does not limit CPU use for empty merge sources); CVE-2026-59869 (js-yaml: YAML merge-key chains can force quadratic CPU consumption); GHSA-5p4m-2wfm-xmqj (JS-YAML: Quadratic CPU consumption in !!omap resolution (3.x and 4.x) — CVE-2026) +1 more

High
Category
Supply Chain
Confidence
80% confidence
Finding

Dependency has known vulnerabilities (CVEs). Using packages with unpatched security flaws exposes the environment to known exploits.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: ws==8.20.0 — 2 advisory(ies): CVE-2026-45736 (ws: Uninitialized memory disclosure); CVE-2026-48779 (ws: Memory exhaustion DoS from tiny fragments and data chunks)

High
Category
Supply Chain
Confidence
80% confidence
Finding

Dependency has known vulnerabilities (CVEs). Using packages with unpatched security flaws exposes the environment to known exploits.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: basic-ftp==5.2.0 — 4 advisory(ies): GHSA-6v7q-wjvx-w8wg (basic-ftp: Incomplete CRLF Injection Protection Allows Arbitrary FTP Command Exe); CVE-2026-39983 (basic-ftp has FTP Command Injection via CRLF); CVE-2026-41324 (basic-ftp vulnerable to denial of service via unbounded memory consumption in Cl) +1 more

High
Category
Supply Chain
Confidence
95% confidence
Finding

The lockfile pins basic-ftp 5.2.0, and the listed advisories indicate command-injection and denial-of-service issues in that package. In this dependency graph it is pulled in transitively via get-uri/pac-proxy-agent/proxy-agent for proxy handling, so exploitation depends on the skill processing attacker-controlled FTP/PAC/proxy inputs, but the vulnerable package is genuinely present.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: extract-zip==2.0.1 — 2 advisory(ies): CVE-2026-19693 (extract-zip allows arbitrary file writes through symlink archive entries); CVE-2026-56876 (extract-zip unvalidated symlink path traversal)

High
Category
Supply Chain
Confidence
96% confidence
Finding

extract-zip 2.0.1 is present and is used by @puppeteer/browsers during browser download/extraction. Archive extraction flaws involving symlinks and path traversal can lead to arbitrary file write if an attacker can influence the downloaded archive or extraction source, which is especially relevant for tooling that fetches browser binaries.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: ip-address==10.1.0 — 2 advisory(ies): CVE-2026-69192 (ip-address: Address4 decodes leading-zero octets as decimal while resolvers deco); CVE-2026-42338 (ip-address has XSS in Address6 HTML-emitting methods)

High
Category
Supply Chain
Confidence
88% confidence
Finding

ip-address 10.1.0 is included transitively through socks, and the advisory set indicates parsing inconsistencies and XSS in HTML-emitting helpers. This is a real vulnerable dependency, but in this skill context the impact is limited unless the application renders attacker-controlled Address6 HTML output or makes security decisions based on ambiguous IP parsing.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: js-yaml==4.1.1 — 4 advisory(ies): CVE-2026-84375 (js-yaml: maxTotalMergeKeys does not limit CPU use for empty merge sources); CVE-2026-59869 (js-yaml: YAML merge-key chains can force quadratic CPU consumption); GHSA-5p4m-2wfm-xmqj (JS-YAML: Quadratic CPU consumption in !!omap resolution (3.x and 4.x) — CVE-2026) +1 more

High
Category
Supply Chain
Confidence
90% confidence
Finding

js-yaml 4.1.1 is present via cosmiconfig and has advisories for quadratic/CPU-exhaustion behavior on crafted YAML content. If the skill loads attacker-controlled YAML configuration, this could enable denial of service; otherwise exposure is more limited, but the vulnerable package is still genuinely included.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: ws==8.20.0 — 2 advisory(ies): CVE-2026-45736 (ws: Uninitialized memory disclosure); CVE-2026-48779 (ws: Memory exhaustion DoS from tiny fragments and data chunks)

High
Category
Supply Chain
Confidence
91% confidence
Finding

ws 8.20.0 is included via puppeteer-core for browser protocol communication, and the advisories describe memory disclosure and memory exhaustion conditions. Because Puppeteer commonly opens WebSocket connections to the browser, a vulnerable ws dependency can matter if an attacker can influence or intercept those connections, though this is not obviously exposed directly by the skill’s poster-design purpose.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill declares an activation condition broad enough to trigger on essentially any request for a vertical promotional image, not just this narrowly defined poster workflow. Overly broad triggers can cause unintended skill invocation, increasing the chance the agent routes unrelated user tasks into this skill, produces irrelevant outputs, or bypasses more appropriate/specialized skills.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The template hard-codes lang="zh-CN", which imposes a specific language/locale choice in the skill output. Under the policy, locale constraints should either be user-selectable or clearly justified as region-specific; this document does not provide such an opt-in or justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The script loads raw HTML into a full browser context with page.setContent(..., { waitUntil: 'networkidle0' }), which allows embedded external resources such as images, fonts, stylesheets, and scripts to trigger outbound network requests during rendering. If the HTML content is attacker-controlled or derived from user input, this can cause server-side request forgery, internal network probing, tracking beacons, or unintended data egress from the environment performing the render.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The file’s docstrings and console messages are written in Chinese, and the script does not offer any language or locale selection. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale constraint is explicitly justified, which is not present here.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The script copies both the generated PNG and the source HTML into a timestamped folder under the user's Downloads directory, which exceeds the stated poster-rendering purpose. Because the HTML may contain user-provided content, prompts, embedded data, or local references, this creates unintended data propagation into a user-visible location and broadens the skill's side effects beyond rendering.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

Copying the input HTML source into Downloads is an unnecessary capability for an image-generation skill and may expose user content, hidden metadata, prompts, local file references, or sensitive embedded information. The risk is elevated because the HTML is the source artifact, not merely the final rendered image, so it can disclose far more than the user expects from poster generation.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
94% confidence
Finding

The dependency version for puppeteer is specified with a caret range (^24.40.0), which allows automatic installation of newer compatible releases rather than a single exact version. This creates supply-chain and reproducibility risk: a future upstream release could introduce a vulnerable or malicious change and the skill would consume it without an explicit review, especially notable here because Puppeteer is a powerful browser automation library.

Content

Scanner excerpt · package.json (reported line 14)May include surrounding context.

json
"license": "ISC",
  "type": "commonjs",
  "dependencies": {
    "puppeteer": "^24.40.0"
  }
}

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

L604-L608 说明 body 设置 padding: 60px 会形成 1080x1680 内容区,且正文为 20-22px;但同一文档前文 L011-L013、L012 规定画布为 1200x1800、上下 150px、左右 120px,内容区为 960x1500,L084-L087 也将正文规格定义为 28-32px 等更大的尺寸。这不是单纯信息缺失,而是文档内部对实现目标给出互相冲突的指令,可能导致技能实际生成结果偏离其宣称的统一版式。

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
88% confidence
Finding

The dependency uses a caret version range, which allows automatic installation of future 24.x releases of Puppeteer. This weakens build reproducibility and increases supply-chain risk because a compromised or breaking upstream release could be pulled in without an explicit review. In this skill, Puppeteer is a browser automation library, so an unsafe update could affect any rendering or automation flow that uses it.

Content

Scanner excerpt · scripts/puppeteer/package.json (reported line 3)May include surrounding context.

json
{
  "dependencies": {
    "puppeteer": "^24.40.0"
  }
}