Back to skill

Security audit

desktop-app-standard

Security checks for vulnerabilities and agentic risk

Overview

ClawScan judged the skill benign, with the main caveat that its purpose-related review is only a note because some files were not fully inspected.

The verdict is benign, but it does not mean risk-free. Consider that Skillspector's analysis was partial: it reports 23 of 37 components scanned and 14 partially inspected, so some evidence is incomplete. Review the uninspected portions of the reported scripts before relying on the skill.

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (52)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 13)May include surrounding context.

md
- scripts/audit.mjs

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 14)May include surrounding context.

md
- scripts/core.mjs

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 15)May include surrounding context.

md
- scripts/app-context.mjs

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 16)May include surrounding context.

md
- scripts/ci-traps.mjs

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 18)May include surrounding context.

md
- scripts/dotnet.mjs

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 19)May include surrounding context.

md
- scripts/review-baseline.mjs

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 20)May include surrounding context.

md
- scripts/checksums.mjs

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 21)May include surrounding context.

md
- scripts/release-assets.mjs

Chaining Abuse

High
Category
Tool Misuse
Confidence
75% confidence
Finding

Tool calls are chained to bypass individual safety checks or escalate capabilities beyond what any single tool call would allow.

Content

Scanner excerpt · templates/workflows/ci.yml (reported line 33)May include surrounding context.

yaml
# Adapt to existing scripts: omit absent lint; do not repeat typecheck if npm test includes it.
      # ---- Tauri only: skipped automatically in an Electron repo ----
      - if: runner.os == 'Linux' && hashFiles('src-tauri/Cargo.toml') != ''
        run: |
          sudo apt-get update
          sudo apt-get install -y libwebkit2gtk-4.1-dev build-essential curl wget file libxdo-dev libssl-dev libayatana-appindicator3-dev librsvg2-dev patchelf xdg-utils
      - if: hashFiles('src-tauri/Cargo.toml') != ''

Chaining Abuse

High
Category
Tool Misuse
Confidence
75% confidence
Finding

Tool calls are chained to bypass individual safety checks or escalate capabilities beyond what any single tool call would allow.

Content

Scanner excerpt · templates/workflows/release-tauri.yml (reported line 56)May include surrounding context.

yaml
}
          NODE
      - run: npm ci                     # Yarn 1: yarn install --frozen-lockfile · Yarn 4: corepack enable + yarn install --immutable
      - run: |
          sudo apt-get update
          sudo apt-get install -y libwebkit2gtk-4.1-dev build-essential curl wget file libxdo-dev libssl-dev libayatana-appindicator3-dev librsvg2-dev patchelf xdg-utils
      - uses: dtolnay/rust-toolchain@stable

Chaining Abuse

High
Category
Tool Misuse
Confidence
75% confidence
Finding

Tool calls are chained to bypass individual safety checks or escalate capabilities beyond what any single tool call would allow.

Content

Scanner excerpt · templates/workflows/release-tauri.yml (reported line 114)May include surrounding context.

yaml
- uses: actions/checkout@v7
        with: { ref: '${{ needs.validate.outputs.commit }}' }
      - if: runner.os == 'Linux'
        run: |
          sudo apt-get update
          sudo apt-get install -y libwebkit2gtk-4.1-dev build-essential curl wget file libxdo-dev libssl-dev libayatana-appindicator3-dev librsvg2-dev patchelf xdg-utils
      - uses: actions/setup-node@v7

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, shell.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Requires English installer and uninstaller interfaces without language-specific user opt-in.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The delivery instructions establish English installation and uninstallation interfaces as a mandatory house-style standard. This imposes a natural-language choice without explicitly asking the user to select it or providing a locale-specific justification.

Content

No source excerpt is available for this finding.

Directs agents to configure an English-only NSIS installer and disable language selection.

Medium
Category
Output Handling
Confidence
99% confidence
Finding

The workflow adaptation instructions explicitly set installerLanguages to [en_US] and displayLanguageSelector to false. This forces a particular interface locale without user opt-in or a documented justification.

Content

No source excerpt is available for this finding.

Mandates English-only installation and uninstallation interfaces regardless of system or application language.

Medium
Category
Output Handling
Confidence
100% confidence
Finding

The instruction requires English for wizard text, custom pages, buttons, and prompts, removes other languages, and disables language selection. It explicitly prevents following the system or application language and requires verification on Chinese Windows, without offering a language choice or explaining a justified locale constraint.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/audit.mjs (reported line 540)May include surrounding context.

js
detail: `插件 ${p} 的 Rust(${rust.map(x => x.resolved).join('/')}) 与 JS(${js.map(x => x.resolved).join('/')}) 版本不一致;不会在构建时失败,而是运行时缺命令/行为错位。`, fix: '同一提交中把两侧升级到同一完整版本。' });
    }
    if (!local || !target || compare(local, target) >= 0) return;
    const url = `https://api.github.com/repos/tauri-apps/plugins-workspace/contents/plugins/${p}/CHANGELOG.md?ref=v2%60;
    const md = await get(url, { raw: true, allow404: true });
    if (md === null) { report.warnings.push(`插件 ${p}: plugins-workspace 中没有 CHANGELOG(第三方插件?),需到其自身仓库核验。`); return; }
    const web = `https://github.com/tauri-apps/plugins-workspace/blob/v2/plugins/${p}/CHANGELOG.md`;

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/audit.mjs (reported line 575)May include surrounding context.

js
}
// Action pins: latest release and major gap. House style pins major tags, so suggestedPin is `@v<latest major>`.
if (!offline) for (const repository of [...new Set(report.actions.map(a => a.repository))]) await collect(`action:${repository}`, async () => {
  const latest = await get(`https://api.github.com/repos/${repository}/releases/latest`, { allow404: true });
  const tags = await get(`https://api.github.com/repos/${repository}/tags?per_page=100%60);
  const latestTag = latest?.tag_name ?? null, latestVer = refVersion(latestTag ?? '');
  const majorTags = tags.filter(t => /^v\d+$/.test(t.name)).map(t => Number(t.name.slice(1)));

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/audit.mjs (reported line 576)May include surrounding context.

js
// Action pins: latest release and major gap. House style pins major tags, so suggestedPin is `@v<latest major>`.
if (!offline) for (const repository of [...new Set(report.actions.map(a => a.repository))]) await collect(`action:${repository}`, async () => {
  const latest = await get(`https://api.github.com/repos/${repository}/releases/latest`, { allow404: true });
  const tags = await get(`https://api.github.com/repos/${repository}/tags?per_page=100%60);
  const latestTag = latest?.tag_name ?? null, latestVer = refVersion(latestTag ?? '');
  const majorTags = tags.filter(t => /^v\d+$/.test(t.name)).map(t => Number(t.name.slice(1)));
  const latestMajor = latestVer?.major ?? (majorTags.length ? Math.max(...majorTags) : null);

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/core.mjs (reported line 181)May include surrounding context.

js
const missing = expectedVersions === null ? null : new Set(expectedVersions);
  const visited = [];
  for (let page = 1; page <= maxPages; page++) {
    const url = `https://api.github.com/repos/${repo}/releases?per_page=100&page=%24%7Bpage%7D%60;
    visited.push(url);
    const data = await get(url);
    if (!Array.isArray(data)) throw new Error('release API 未返回数组');

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · tests/cli.test.mjs (reported line 666)May include surrounding context.

js
'https://registry.npmjs.org/electron-updater': { 'dist-tags': { latest: '7.0.0', next: '6.8.20' }, versions: { '6.8.9': {}, '6.8.10': {}, '6.8.20': {}, '6.8.12-beta.1': {}, '7.0.0': {} } },
      }[url];
      if (data) return { ok: true, json: async () => data };
      if (url.startsWith('https://api.github.com/repos/electron-userland/electron-builder/releases')) return { ok: true, json: async () => [] };
      throw new Error('unexpected request ' + url);
    };`);
    const r = spawnSync(process.execPath, ['--import', pathToFileURL(mock).href,

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · tests/core.test.mjs (reported line 63)May include surrounding context.

js
test('GitHub credential is not sent to npm/crates/official Electron index', async () => {
  const seen = [];
  const fetchImpl = async (u, opts) => { seen.push({ u, opts }); return { ok: true, status: 200, json: async () => ({}) }; };
  for (const url of ['https://api.github.com/repos/a/b/tags', 'https://registry.npmjs.org/electron/latest', 'https://crates.io/api/v1/crates/tauri', 'https://releases.electronjs.org/releases.json', 'https://builds.dotnet.microsoft.com/dotnet/release-metadata/10.0/releases.json', 'https://raw.githubusercontent.com/nodejs/Release/main/schedule.json']) await requestJson(url, { token: 'test-secret', fetchImpl });
  assert.equal(seen[0].opts.headers.Authorization, 'Bearer test-secret');
  for (const x of seen.slice(1)) assert.equal(x.opts.headers.Authorization, undefined);
  assert.ok(seen.every(x => x.opts.redirect === 'error'));

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · tests/core.test.mjs (reported line 71)May include surrounding context.

js
test('untrusted destinations and rate-limit responses fail explicitly', async () => {
  await assert.rejects(requestJson('https://example.com/x'), /白名单/);
  await assert.rejects(requestJson('http://api.github.com/x'), /白名单/);
  await assert.rejects(requestJson('https://api.github.com/x', { fetchImpl: async () => ({ ok: false, status: 429 }) }), /未知/);
});
test('5xx retries once', async () => {
  let count = 0;

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · tests/core.test.mjs (reported line 75)May include surrounding context.

js
});
test('5xx retries once', async () => {
  let count = 0;
  const x = await requestJson('https://api.github.com/x', { fetchImpl: async () => (++count === 1 ? { ok: false, status: 503 } : { ok: true, status: 200, json: async () => ({ value: 1 }) }) });
  assert.equal(count, 2); assert.equal(x.value, 1);
});
const release = v => ({ tag_name: 'v' + v, body: '`app.test` changed', html_url: 'https://github.com/test/test/releases/v' + v, published_at: '2026-01-01T00:00:00Z', prerelease: false, draft: false });

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · tests/release-assets.test.mjs (reported line 128)May include surrounding context.

js
verifyFeed(data.feed, {...options, preferNsis: true});
    const entry = data.feed.platforms['windows-x86_64'];
    verify({...data.feed, platforms: {'windows-x86_64': {...entry, url: `https://github.com/org/app/releases/download/v1.2.3/${data.names[0]}`}}});
    for (const url of ['https://api.github.com/repos/org/app/releases/assets/999',
      `https://github.com/other/app/releases/download/v1.2.3/${data.names[0]}`,
      `https://github.com/org/app/releases/download/v0.0.1/${data.names[0]}`]) {
      assert.throws(() => verify({...data.feed, platforms: {'windows-x86_64': {...entry, url}}}));

MCP server referenced without pinned version: 'npx electron'.

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.