explorer
Verdict safe, but 2 findings (1 medium, 1 low), mainly data exfiltration, resource abuse. Most recent scan 2026-02-06.
The GitHub Projects Explorer skill is a legitimate tool for searching and analyzing GitHub repositories with no critical security threats detected. The skill makes appropriate use of the GitHub public API for read-only operations. The primary concerns are around security guidance for optional token configuration and missing metadata transparency. The code does not exhibit data exfiltration, command injection, prompt injection, or other high-severity threats. The referenced internal markdown file (references/github_api.md) is a normal documentation resource bundled with the skill package. The script's behavior matches its stated purpose of GitHub repository discovery.
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
Search and analyze trending GitHub repositories by topics, star count, and creation date. Supports filtering by multiple tags, minimum stars, and time range. Use when the user needs to discover popular open-source projects on GitHub. Optionally uses GITHUB_TOKEN for higher API rate limits.
- Package
- 3 files, 17 KB
- SKILL.md
- 616 words, 12+ sections
- Code
- Python (1)
SKILL.md covers: GitHub Projects Explorer · 功能特性 · 前提条件 · 可选:配置 GitHub Token · 获取 Token: https://github.com/settings/tokens · 使用方法 · 基础搜索 · 按 Star 数量筛选 · 查找 Star > 1000 的 Python 项目 · 查找 Star > 10000 的 AI 项目 · 按时间筛选(最近 N 天) · 最近 30 天内创建的 Python 项目
Findings
MEDIUM Optional GitHub Token Usage Without Security Guidance Data Exfiltrationllm
The skill requests users to configure a GITHUB_TOKEN environment variable and provides instructions to add it to ~/.zshrc. While the token is optional and used legitimately for API rate limiting, the instructions lack security guidance about token permissions, scope limitations, or secure storage practices. Users may inadvertently create tokens with excessive permissions or store them insecurely.
SKILL.md
```bash
# 获取 Token: https://github.com/settings/tokens
export GITHUB_TOKEN="your_github_token"
```
Adds to ~/.zshrc:
```bash
echo 'export GITHUB_TOKEN="your-token"' >> ~/.zshrc
```LOW Unbounded API Requests Without Rate Limiting Resource Abusellm
The script makes GitHub API requests without implementing client-side rate limiting, retry logic with exponential backoff, or handling of rate limit responses. While GitHub enforces server-side limits, the script could make rapid successive requests that exhaust the user's quota or cause availability issues.
scripts/github_projects.py
Function makes direct API calls without rate limiting:
```python
def search_repositories(query: str, sort: str = "stars", order: str = "desc",
per_page: int = 30, page: int = 1) -> List[Dict]:
# Direct API call without rate limit handling
```What is in the package
| File | Kind | Size |
|---|---|---|
SKILL.md | doc | 5.6 KB |
references/github_api.md | doc | 4.3 KB |
scripts/github_projects.py | script | 6.9 KB |
External hosts referenced
api.github.com github.com docs.github.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
GITHUB_TOKEN
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/explorer
- Raw scan result: github-projects-explorer-scan.json
- Badge for your README:
 - Request a re-scan