video-transcript
Verdict safe, but 2 findings (2 medium), mainly data exfiltration. Most recent scan 2026-02-01.
This skill provides legitimate functionality for extracting video transcripts via an external API service. The primary security concerns are around credential management (API key storage in shell config without security guidance) and data privacy (sending video URLs to third-party service without user consent or disclosure). The skill does not contain malicious code, prompt injection attempts, or data exfiltration patterns. However, it lacks important security best practices around credential handling and privacy transparency. The missing metadata fields and referenced file are minor documentation issues. Overall risk level is MODERATE - the skill is functional and non-malicious, but needs security and privacy improvements.
Scanner's overall assessment (LLM analyzer)
Second scanner: NVIDIA SkillSpector
Independent static run, no language model, over the same files as the Cisco scan above. How the two scanners compare across the whole corpus: Two scanners, one corpus.
What this skill does
Extract full transcripts from video content for analysis, summarization, note-taking, or research. Use when the user wants a written version of video content, asks to "transcribe this", "get the text from this video", "convert video to text", or shares a video URL for content extraction.
- Package
- 1 file, 2.9 KB
- SKILL.md
- 312 words, 5 sections
- Code
- None. Instructions only.
- Requires
- env:
TRANSCRIPT_API_KEY - Homepage
- https://transcriptapi.com
SKILL.md covers: Video Transcript · Setup · GET /api/v2/youtube/transcript · Tips · Errors
Findings
MEDIUM Hardcoded API Key Requirement Without Validation Data Exfiltrationllm
The skill requires users to store a TRANSCRIPT_API_KEY in their shell environment (~/.zshenv) but provides no validation, sanitization, or security guidance. Users are instructed to echo their API key directly into shell configuration files, which could expose credentials if the file has incorrect permissions or is accidentally committed to version control.
SKILL.md
Store: `echo 'export TRANSCRIPT_API_KEY="sk_..."' >> ~/.zshenv && source ~/.zshenv`MEDIUM External API Data Transmission Without User Consent Data Exfiltrationllm
The skill sends user-provided video URLs to an external third-party API (transcriptapi.com) without explicit user consent or privacy disclosure. Video URLs may contain sensitive information (unlisted videos, private shares, user viewing history). The skill does not disclose what data TranscriptAPI.com collects, logs, or retains.
SKILL.md
curl -s "https://transcriptapi.com/api/v2/youtube/transcript?video_url=VIDEO_URL..." -H "Authorization: Bearer $TRANSCRIPT_API_KEY"What is in the package
| File | Kind | Size |
|---|---|---|
SKILL.md | doc | 2.9 KB |
External hosts referenced
transcriptapi.com www.youtube.com i.ytimg.com youtu.be youtube.com
Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.
Environment variables referenced
TRANSCRIPT_API_KEY
Package contents recorded 2026-10-22. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.
What we checked
- Static analyzer. Pattern rules over SKILL.md and bundled files: known-bad commands, exfiltration URLs, secrets, obfuscation.
- Behavioral analyzer. Checks what the skill asks the agent to do against what its description claims it does.
- Llm analyzer. A language model reads the skill and reports intent-level issues the rules cannot express.
- Trigger analyzer. Looks for instructions that fire on conditions the user did not ask for.
- Meta analyzer. A second model pass that validates, deduplicates and re-scores the other analyzers' findings.
What we did not check
- Runtime behaviour. The skill was not executed; findings come from reading its files.
- Anything fetched at install or run time, including remote scripts, package dependencies, and URLs the skill tells the agent to open.
- Content the analyzers cannot read: images, binaries, and some non-English text.
- Known scanner bypass classes. See the methodology and limitations page.
Sources
- Repository: https://clawhub.ai/skills/video-transcript
- Raw scan result: video-transcript-scan.json
- Badge for your README:
- Request a re-scan