youtube-channels
Verdict safe, but 1 finding (1 medium), mainly data exfiltration. Most recent scan 2026-02-01.
The youtube-channels skill is a relatively safe wrapper around the TranscriptAPI.com service with no critical security vulnerabilities. The skill contains no executable scripts, no data exfiltration code, no prompt injection attempts, and no malicious behavior. The primary concerns are around credential management practices in the setup instructions (MEDIUM severity) and missing optional metadata transparency (LOW severity). The skill accurately describes its functionality and doesn't attempt to deceive users about its purpose. The external API dependency and credit-based pricing model should be more prominently disclosed, but this represents a transparency issue rather than a security threat. No executable code means no command injection, eval/exec risks, or tool exploitation vectors. Overall, this skill demonstrates good security hygiene with room for improvement in credential handling guidance and metadata completeness.
Scanner's overall assessment (LLM analyzer)
Second scanner: NVIDIA SkillSpector
SkillSpector found no issues but could not fully inspect the package, usually because SKILL.md refers to a file that is not present, so it fails closed to CAUTION instead of SAFE. Read it as "not fully checked", not as a finding. How the two scanners compare across the whole corpus: Two scanners, one corpus.
What this skill does
Work with YouTube channels — resolve handles to IDs, browse uploads, get latest videos, search within channels. Use when the user asks about a specific channel, wants to see recent uploads, or says "what has X posted lately", "latest from MKBHD", "show me their channel", "list channel videos", "browse channel uploads".
- Package
- 1 file, 5.3 KB
- SKILL.md
- 505 words, 12+ sections
- Code
- None. Instructions only.
- Requires
- env:
TRANSCRIPT_API_KEY - Homepage
- https://transcriptapi.com
SKILL.md covers: YouTube Channels · Setup · GET /api/v2/youtube/channel/resolve — FREE · GET /api/v2/youtube/channel/latest — FREE · GET /api/v2/youtube/channel/videos — 1 credit/page · First page · Next pages · GET /api/v2/youtube/channel/search — 1 credit · Typical workflow · 1. Resolve handle to ID (free) · 2. Check latest uploads (free) · 3. Get transcript of recent video
Findings
MEDIUM Hardcoded API Key Exposure Risk in Setup Instructions Data Exfiltrationllm
The SKILL.md instructions guide users to store their TRANSCRIPT_API_KEY in ~/.zshenv using echo append, which could lead to accidental exposure or misconfiguration. While the skill itself doesn't contain hardcoded secrets, the setup pattern encourages storing sensitive credentials in shell configuration files that may be inadvertently committed to version control or exposed through other means. The instructions also demonstrate the API key in curl commands, which could lead to shell history exposure.
SKILL.md
Store: `echo 'export TRANSCRIPT_API_KEY="sk_..."' >> ~/.zshenv && source ~/.zshenv`
Auth: `-H "Authorization: Bearer $TRANSCRIPT_API_KEY"`What is in the package
| File | Kind | Size |
|---|---|---|
SKILL.md | doc | 5.3 KB |
External hosts referenced
transcriptapi.com www.youtube.com i1.ytimg.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/youtube-channels
- Raw scan result: youtube-channels-scan.json
- Badge for your README:
- Request a re-scan