doc-speedreader

Fail

Audited by Snyk on Jul 22, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E005: Suspicious download URL detected in skill instructions.

  • Suspicious download URL detected (high risk: 0.80). Two URLs are high-risk: one is an internal/private IP admin endpoint (SSRF/credential attack surface) and the other is an explicit "evil" remote URL shown in the safety docs as a dangerous example — both could be used to deliver or trigger malware/attacks if fetched/executed.

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

  • Third-party content exposure detected (high risk: 0.85). outsider-authored free text is the user-provided document content (from local file paths or fetched remote URLs) that MarkItDown converts into readable Markdown and then scripts/convert.sh writes to stdout (sys.stdout.write(result.text_content)), which the runtime workflow would feed back into the LLM context for summarization.

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

  • Potentially malicious external URL detected (high risk: 0.80). The skill may run scripts/install.sh at runtime (SKILL.md:54-57) which calls pip to fetch and install the required markitdown package from PyPI using PIP_INDEX "https://pypi.org/simple" (and fallback "https://mirrors.aliyun.com/pypi/simple/"), so remote code is fetched during runtime and will be executed as a required dependency.

Issues (3)

E005
CRITICAL

Suspicious download URL detected in skill instructions.

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
Jul 22, 2026, 03:48 AM
Issues
3