githunt

SAFE max severity MEDIUMSkillSpector Safe 17 registry clawhub.ai last scanned 2026-02-06 1 scan

Verdict safe, but 2 findings (1 medium, 1 low), mainly data exfiltration. Most recent scan 2026-02-06.

The githunt skill is a developer search tool that queries an external API (githunt.ai) to find GitHub developers by location, technology stack, and role. The skill is MODERATELY SAFE with some privacy and transparency concerns. Primary issues: (1) Automated data transmission to third-party API without explicit user consent or privacy warnings, (2) Missing manifest metadata reducing transparency, (3) Lack of API endpoint validation and rate limiting. The skill does not contain malicious code, prompt injection, credential theft, or command injection vulnerabilities. The external API calls are legitimate for the stated purpose, but users should be informed about data sharing. The bash scripts are straightforward with no obfuscation or hidden functionality. Recommended for use with minor improvements to user consent and metadata completeness.
Scanner's overall assessment (LLM analyzer)

Findings2

Critical0

High0

Medium1

Low1

Analyzersstatic behavioral llm trigger meta

Second scanner: NVIDIA SkillSpector

VerdictSafe 17

Score17 / 100

Issues25

CategoriesData Exfiltration MCP Least Privilege

Version2.12.0 (static)

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

Find and rank GitHub developers by location, technology, and role. Search for candidates, get scored profiles with tech stack matches, activity, and contact info.

Package: 6 files, 17 KB
SKILL.md: 637 words, 12+ sections
Code: Shell (2)
Version: 1.0.0

SKILL.md covers: GitHunt - GitHub Developer Discovery · When to Use · API Endpoints · Search Developers (Streaming) - Main Endpoint · Supported Roles · Rank Single User · Stream Response Format · User Data Fields · Free vs Paid · 💰 Get Full Report · Usage Examples · Find React Developers in Berlin (Streaming)

Findings

MEDIUM Potential Data Exfiltration via Third-Party API Without User Consent Data Exfiltrationllm

The skill makes automated network requests to an external third-party API (api.githunt.ai) that collects and transmits user search queries including location data, technology preferences, and role information. While the API itself appears legitimate for developer search purposes, the skill does not implement any user consent mechanism or data privacy warnings before sending potentially sensitive search criteria to external servers. Users may unknowingly share their hiring intentions, technology interests, and geographic targeting information with a third-party service.

scripts/githunt-search.sh

curl -s -N -X POST "$API_URL/rank/users/stream" \
  -H "Content-Type: application/json" \
  -H "Accept: text/event-stream" \
  -d "$payload"

Remediation: 1. Add explicit user consent prompt before making API calls to external services. 2. Include privacy notice in SKILL.md about data transmission to githunt.ai. 3. Implement local caching to minimize repeated API calls. 4. Add option for users to review payload before transmission. 5. Document what data is sent to third-party services in the manifest.

LOW Hardcoded API Endpoint Without Verification Data Exfiltrationllm

The skill uses a hardcoded external API endpoint (https://api.githunt.ai/v1) with a fallback mechanism via environment variable GITHUNT_API_URL. While environment variable override is good practice, there is no validation of the API endpoint URL, certificate pinning, or integrity checks. This could allow an attacker who controls the environment to redirect API calls to a malicious endpoint for data interception.

scripts/githunt-search.sh:8

API_URL="${GITHUNT_API_URL:-https://api.githunt.ai/v1}"

Remediation: 1. Implement URL validation to ensure only HTTPS endpoints are used. 2. Add certificate pinning or domain allowlisting for production API endpoints. 3. Validate API responses for expected structure before processing. 4. Consider adding API endpoint verification in documentation.

What is in the package

File	Kind	Size
`SKILL.md`	doc	4.9 KB
`_meta.json`	config	528 B
`scripts/githunt-search.sh`	script	4.2 KB
`skill/SKILL.md`	doc	5.3 KB
`skill/_meta.json`	config	528 B
`skill/scripts/githunt-search.sh`	script	2.0 KB

External hosts referenced

api.githunt.ai githunt.ai github.com avatars.githubusercontent.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

API_URL GITHUNT_API_URL

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/githunt
Raw scan result: githunt-scan.json
Badge for your README: ![skillscan.io](https://skillscan.io/badge/clawhub-ai-githunt.svg)
Request a re-scan