Back to skill

Security audit

sketchy

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious with medium confidence because it reports a risk of exposing local secrets in generated files, unsafe privileged installation advice, and incomplete inspection of some components.

Consider the reported secret-exposure risk and privileged installation advice before installing. Scanner coverage is partial: 39 of 56 components were fully inspected and 17 were partially inspected, so the results do not establish that the remaining behavior is safe. The verdict is suspicious; it does not conclude that the author is malicious.

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/build.mjs
Finding

Shared themes can embed unrelated local secrets into deliverable HTML

Content
View full analysis

Impact

A generated HTML file can contain unrelated local secrets in its source without displaying them. Anyone receiving that file can decode the embedded bytes. No attacker-controlled network endpoint is required.

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (34)

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 60)May include surrounding context.

md
10. **Self-contained output.** The final file runs from `file://` with no network. Build it with `scripts/build.mjs`. Fonts are bundled (SIL OFL).

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 83)May include surrounding context.

md
node scripts/build.mjs <file.html> # one self-contained file in <folder>/out/

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 73)May include surrounding context.

md
Run `node scripts/deal.mjs --brief "<one line>"`. It deals paper and theme, cover color, the two leading pencils, a motif and the opening page, and it avoids th

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 79)May include surrounding context.

md
Run `node scripts/new.mjs <deck|explainer|report|chart|board|postcard> <out-folder> [--theme <name>] [--title "..."]`. It copies the template and the engine int

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 84)May include surrounding context.

md
node scripts/lint.mjs <out/file.html> # overflow, collisions, tiny type, contrast, words per slide, placeholders, network

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 85)May include surrounding context.

md
node scripts/shoot.mjs <out/file.html> # every slide or page at 1920x1080 and 390x844, plus contact-sheet.png

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 94)May include surrounding context.

md
node scripts/export-pdf.mjs <out/file.html>

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 95)May include surrounding context.

md
node scripts/export-pptx.mjs <out/file.html> --mode faithful # or editable; see references/export.md

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 96)May include surrounding context.

md
node scripts/render-video.mjs <out/file.html> --fps 30 # MP4 with generated music; needs FFmpeg

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 114)May include surrounding context.

md
Not done: PPTX (not asked), video (FFmpeg missing; `node scripts/doctor.mjs` prints the install line)

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 161)May include surrounding context.

md
eed Node 18+ and Playwright 1.63.0 with Chromium; video needs FFmpeg. Run `node scripts/doctor.mjs` first when anything fails: it prints the exact install line

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 131)May include surrounding context.

md
| `engine/sketchy.css` | tokens (day and night), paper, type, sheets, buttons, cover, stamps, slips, tape, postcards, gauges, motion |

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 132)May include surrounding context.

md
| `engine/sketchy.js` | the pencil: seeded geometry, pen API, boil, decorators (`data-sk`), postal objects, paper, `Sketchy.mount()` |

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 133)May include surrounding context.

md
| `engine/sketches.js` | the sketch catalog, more than 100 small drawings of things |

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 134)May include surrounding context.

md
| `engine/sketchy-charts.js` | hand-drawn charts with honest data (`<figure class="sk-chart">`) |

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 135)May include surrounding context.

md
| `engine/sketchy-deck.js` | the slide runtime: stage, page turn, steps, counters, overview, notes, print and video modes |

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 136)May include surrounding context.

md
| `engine/sketchy-sound.js` | procedural sound effects and generative music |

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

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.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

The instruction to read this reference for “anything with numbers, money, plans, results or decisions” is overly broad.

Medium
Category
Trigger Abuse
Confidence
88% confidence
Finding

This activation condition covers routine calculations, planning, and decision requests without limiting use to creating Sketchy business visuals. The stationery guidance explains what to produce after activation but does not explicitly exclude unrelated tasks.

Content

No source excerpt is available for this finding.

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

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.

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

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.

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

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.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
88% confidence
Finding

The issue is not sudo alone, but recommending root execution of an npm CLI selected through npx resolution. Following this advice can elevate compromised project dependency code or an implicitly downloaded package to root, although the diagnostic itself merely prints the command.

Content

Scanner excerpt · scripts/_lib.mjs (reported line 196)May include surrounding context.

js
} catch (e) {
    const msg = String(e && e.message ? e.message : e);
    if (/Executable doesn't exist|playwright install/i.test(msg)) die('Chromium is not installed for Playwright', chromiumInstallFix());
    if (/missing dependencies|shared librar|libnss|libatk/i.test(msg)) die('Chromium cannot start: system libraries are missing', 'Linux: sudo npx playwright install-deps chromium');
    die(`Chromium failed to start: ${msg.split('\n')[0]}`, 'node doctor.mjs');
  }
}

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

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

Although this is not an MCP server, the suggested repair runs an implicitly resolved npm CLI through sudo without an explicit package version or verified CLI path. If followed in a compromised project or where the intended package is not resolved, it can execute attacker-controlled or newly fetched JavaScript as root; the script only prints this advice and does not execute it automatically.

Content

No source excerpt is available for this finding.