generate-slides-retro-simple

Fail

Audited by Snyk on Aug 18, 2026

Risk Level: CRITICAL
Full Analysis

HIGH W007: Insecure credential handling detected in skill instructions.

  • Insecure credential handling detected (high risk: 1.00). De skill vraagt expliciet de gebruiker het GitHub Personal Access Token in de chat te plakken en toont een export/verificatielijn, waardoor het model het geheime token in zijn context krijgt en het (per ongeluk) woordelijk kan moeten doorgeven.

CRITICAL E006: Malicious code pattern detected in skill scripts.

  • Malicious code pattern detected (high risk: 0.90). The skill explicitly instructs the user to paste a GitHub Personal Access Token into the chat and export/use it for API calls, which is a clear credential-exfiltration vector (sensitive token disclosure to an external/chat endpoint).

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (medium risk: 0.30). De workflow leest outsider-authored vrije tekst uit GitHub door in Stap 3 via de CEDA Board GraphQL query de project item content (Issue) inclusief body op te halen, waarna deze body wordt gebruikt voor “scope-analyse bij Pitches”.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 0.80). The skill instructs cloning and using https://github.com/cedanl/clidev-presentaties.git at runtime (git clone + npm install + npx slidev), which fetches remote code and themes that are then installed/executed and required to generate the slides.

Issues (4)

W007
HIGH

Insecure credential handling detected in skill instructions.

E006
CRITICAL

Malicious code pattern detected in skill scripts.

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

W012
MEDIUM

Unverifiable external dependency detected (runtime URL that controls agent).

Audit Metadata
Risk Level
CRITICAL
Analyzed
Aug 18, 2026, 09:51 PM
Issues
4