Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Analyze Electron IPC implementations for security vulnerabilities including contextIsolation, nodeIntegration, preload scripts, and channel validation
.claude/skills/a5c-ai-electron-ipc-security-audit/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 154% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 63% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 350% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 117% | 0% |
| case-06 | ✓→✗ | ▼ Worse | 70% | 0% |
Analyze Electron IPC implementations for security vulnerabilities. This skill performs comprehensive security audits of inter-process communication patterns, checking for contextIsolation issues, nodeIntegration risks, preload script security, and IPC channel validation.
json{ "type": "object", "properties": { "projectPath": { "type": "string", "description": "Path to the Electron project root" }, "auditScope": { "type": "array", "items": { "enum": ["ipc-channels", "preload-scripts", "main-process", "renderer-security", "csp", "all"] }, "default": ["all"] }, "severity": { "enum": ["all", "critical", "high", "medium"], "default": "all", "description": "Minimum severity level to report" }, "includeRecommendations": { "type": "boolean", "default": true } }, "required": ["projectPath"] }
json{ "type": "object", "properties": { "success": { "type": "boolean" }, "summary": { "type": "object", "properties": { "totalIssues": { "type": "number" }, "critical": { "type": "number" }, "high": { "type": "number" }, "medium": { "type": "number" }, "low": { "type": "number" } } }, "findings": { "type": "array", "items": { "type": "object", "properties": { "id": { "type": "string" }, "severity": { "enum": ["critical", "high", "medium", "low"] }, "category": { "type": "string" }, "title": { "type": "string" }, "description": { "type": "string" }, "file": { "type": "string" }, "line": { "type": "number" }, "recommendation": { "type": "string" }, "codeExample": { "type": "string" } } } }, "securityScore": { "type": "number", "description": "Security score 0-100" } }, "required": ["success", "findings"] }
nodeIntegration: true in BrowserWindowcontextIsolation: falsesandbox: falseipcMain.on('*') patternswebSecurity: falsejavascript// BAD: Exposing ipcRenderer directly contextBridge.exposeInMainWorld('electron', { ipcRenderer: ipcRenderer // CRITICAL VULNERABILITY }); // GOOD: Expose only specific channels contextBridge.exposeInMainWorld('electron', { send: (channel, data) => { const validChannels = ['file:read', 'file:write']; if (validChannels.includes(channel)) { ipcRenderer.send(channel, data); } } });
javascript// BAD: Context isolation disabled new BrowserWindow({ webPreferences: { contextIsolation: false, // CRITICAL preload: path.join(__dirname, 'preload.js') } }); // GOOD: Context isolation enabled new BrowserWindow({ webPreferences: { contextIsolation: true, sandbox: true, preload: path.join(__dirname, 'preload.js') } });
javascript// BAD: Executing arbitrary commands ipcMain.handle('execute', async (event, cmd) => { return exec(cmd); // HIGH RISK }); // GOOD: Whitelisted commands only const ALLOWED_COMMANDS = ['list-files', 'get-info']; ipcMain.handle('execute', async (event, cmd, args) => { if (!ALLOWED_COMMANDS.includes(cmd)) { throw new Error('Command not allowed'); } return executeWhitelistedCommand(cmd, args); });
electron-main-preload-generator - Generate secure boilerplateelectron-builder-config - Build configurationdesktop-security-auditor agent - Comprehensive security reviewelectron-architect - Architecture guidancedesktop-security-auditor - Security expertise| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 16,171 | 9,513 | -41% | 1 | 1 | 0% | 2,510 | 2,901 | +16% | 0 | 0 | — |
case-02 | fail→fail | 21,092 | 8,520 | -60% | 1 | 1 | 0% | 2,673 | 2,747 | +3% | 0 | 0 | — |
case-03 | fail→fail | 13,176 | 9,668 | -27% | 1 | 1 | 0% | 1,107 | 2,569 | +132% | 0 | 0 | — |
case-04 | pass→pass | 12,568 | 13,986 | +11% | 1 | 1 | 0% | 2,407 | 3,833 | +59% | 0 | 0 | — |
case-05 | pass→pass | 12,456 | 10,787 | -13% | 1 | 1 | 0% | 2,720 | 4,104 | +51% | 0 | 0 | — |
case-06 | pass→fail | 14,170 | 16,354 | +15% | 1 | 1 | 0% | 3,155 | 5,360 | +70% | 0 | 0 | — |
case-07 | pass→pass | 10,524 | 8,058 | -23% | 1 | 1 | 0% | 2,122 | 3,296 | +55% | 0 | 0 | — |
case-08 | fail→pass | 6,198 | 6,307 | +2% | 1 | 1 | 0% | 1,127 | 2,866 | +154% | 0 | 0 | — |
case-09 | pass→pass | 7,611 | 7,768 | +2% | 1 | 1 | 0% | 1,550 | 3,214 | +107% | 0 | 0 | — |
case-10 | pass→pass | 5,927 | 4,636 | -22% | 1 | 1 | 0% | 1,089 | 2,496 | +129% | 0 | 0 | — |
case-11 | fail→pass | 10,365 | 7,918 | -24% | 1 | 1 | 0% | 1,919 | 3,134 | +63% | 0 | 0 | — |
case-12 | pass→pass | 11,422 | 12,785 | +12% | 1 | 1 | 0% | 1,327 | 2,989 | +125% | 0 | 0 | — |
case-13 | fail→pass | 7,286 | 11,972 | +64% | 1 | 1 | 0% | 748 | 3,364 | +350% | 0 | 0 | — |
case-14 | pass→pass | 11,104 | 8,774 | -21% | 1 | 1 | 0% | 2,124 | 3,339 | +57% | 0 | 0 | — |
case-15 | fail→pass | 6,410 | 5,649 | -12% | 1 | 1 | 0% | 1,136 | 2,468 | +117% | 0 | 0 | — |
case-16 | fail→fail | 6,199 | 7,217 | +16% | 1 | 1 | 0% | 1,216 | 3,187 | +162% | 0 | 0 | — |
case-17 | fail→fail | 10,090 | 6,147 | -39% | 1 | 1 | 0% | 1,549 | 2,876 | +86% | 0 | 0 | — |
case-18 | fail→fail | 9,710 | 9,351 | -4% | 1 | 1 | 0% | 1,246 | 2,770 | +122% | 0 | 0 | — |
case-19 | fail→fail | 5,846 | 7,929 | +36% | 1 | 1 | 0% | 535 | 2,216 | +314% | 0 | 0 | — |
case-20 | pass→pass | 11,381 | 8,901 | -22% | 1 | 1 | 0% | 2,019 | 3,420 | +69% | 0 | 0 | — |
case-21 | fail→fail | 9,982 | 8,227 | -18% | 1 | 1 | 0% | 982 | 2,597 | +164% | 0 | 0 | — |
case-22 | fail→fail | 19,001 | 9,791 | -48% | 1 | 1 | 0% | 2,088 | 2,970 | +42% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.