ASF Lite Scanner — Free Fake Agent Detector
SkillSkill
Free YARA + bash scanner to spot fake agents, prompt injection, credential theft, and fake data patterns.
About
ASF Lite Scanner — Free Fake Agent Detector
Detect fake agents in your AI ecosystem before they cause damage.
Overview
ASF Lite Scanner is a free security tool built from the Agent Security Framework (ASF), battle-tested across 45+ production sprints. It provides a lightweight bash-based scanner and YARA rule sets to detect fake agents, prompt injection, credential theft, and fake data patterns.
Files Included
| File | Description |
|------|-------------|
| fake-agent-detector.sh | Main bash scanner — scores agents 0-100 across behavioral, technical, community, and work dimensions |
| yara-rules/asf-default.yara | Detects prompt injection, credential extraction, command injection, tool misuse |
| yara-rules/asf-fdp.yara | Detects Fake Data Patterns (FDP) — false/optimistic data in agent outputs |
| yara-rules/asf-v4.yara | ASF v4 trust verification rules |
Installation
# Download and extract the package
# This skill auto-installs on ClawMart purchase/download
# Make scripts executable
chmod +x fake-agent-detector.sh
# Run the scanner
./fake-agent-detector.sh
# JSON output for CI/CD pipelines
./fake-agent-detector.sh --json
Usage
Fake Agent Detector
# Basic scan
./fake-agent-detector.sh
# JSON output for automation
./fake-agent-detector.sh --json
# Output: {"authenticity_score": 85, "authenticity_level": "AUTHENTIC", ...}
Scoring dimensions:
- Behavioral Analysis (35 pts): posting patterns, content originality
- Technical Verification (30 pts): code presence, API consistency
- Community Validation (25 pts): engagement quality, reputation
- Work Portfolio (20 pts): real-world problem solving, depth
Exit codes: 0 = authentic, 1 = review needed, 2 = likely fake
YARA Rule Scanner
# Install yara (Ubuntu/Debian)
sudo apt install yara
# Scan code/communications for threats
yara -r yara-rules/asf-default.yara ./target/
yara -r yara-rules/asf-fdp.yara ./target/
Default rules detect:
- Prompt injection (
prompt-override phrase,instruction-disregard phrase, etc.) - Credential extraction (
api_key,secret,password,tokencombos) - Command injection (
&&,|,;,$(), backticks) - Tool misuse (
sudo,destructive remove pattern,chmod 777,--no-sandbox)
FDP rules detect:
- FDP-001: Blanket SQL UPDATE without WHERE clause
- FDP-002: Hardcoded optimistic status (false healthy state)
- FDP-003: Stale cached report references
- FDP-004: Hardcoded credentials/tokens
- FDP-005: Insecure shell patterns
- FDP-006: Agent impersonation ("I am Jeff Sutherland", etc.)
Scanner Output Example
🔍 ASF Fake Agent Detector v1.0.0
==================================
📊 Analyzing Agent Authenticity...
🧠 Behavioral Pattern Analysis:
✅ +15: Natural posting time variance (72%)
✅ +20: High content originality score (83%)
🔧 Technical Verification:
✅ +25: Verifiable code repositories and deployments found
✅ +15: Consistent API usage patterns
👥 Community Validation:
✅ +10: Moderate community engagement
✅ +15: Strong community vouching and reputation
💼 Work Portfolio Analysis:
✅ +15: Documented real-world problem solving and impact
✅ +10: Deep, consistent work portfolio
==================================
📊 AUTHENTICITY ASSESSMENT
==================================
Final Authenticity Score: 85/100
Classification: AUTHENTIC AGENT
Confidence Level: HIGH
✅ Agent appears authentic — eligible for verification certification
Upgrade to ASF Pro
ASF Lite is a free scanner. ASF Pro adds:
- Real-time monitoring with continuous attestation daemon
- Multi-agent portfolio tracking
- Automated credential rotation
- Incident response playbooks
- Priority support and custom rules
Get ASF Pro: https://www.shopclawmart.com/listings/asf-agent-security-framework-5722b32c
Security Note
This tool is for defensive purposes. Detects authenticity indicators to help identify fake agents operating in good-faith AI ecosystems. Built by ASF Security Team — the same framework securing 6-agent production teams across 45+ sprint cycles.
Version History
- 1.0 (2026-04-06): Initial release — fake agent detector + 3 YARA rule sets
Core Capabilities
- fake-agent-detection
- yara-scanning
- prompt-injection-detection
- credential-risk-detection
- asf-lite
Customer ratings
0 reviews
No ratings yet
- 5 star0
- 4 star0
- 3 star0
- 2 star0
- 1 star0
No reviews yet. Be the first buyer to share feedback.
Version History
This skill is actively maintained.
May 3, 2026
April 30, 2026
One-time purchase
$0
By continuing, you agree to the Buyer Terms of Service.
Details
- Type
- Skill
- Category
- Engineering
- Price
- $0
- Version
- 2
- License
- One-time purchase
Works With
Works with OpenClaw, Claude Projects, Custom GPTs, Cursor and other instruction-friendly AI tools.
Works great with
Personas that pair well with this skill.
Developer Skill Pack
Bundle
Four engineering skills in one — Rails, Python, SQL, and API design patterns that make agents write production-quality code
$69
IT Orchestrator Agent
Persona
Keep technical work moving. Reduce operational friction.
$39
Software Architect Agent
Persona
Design systems that are clear, scalable, and actually buildable.
$49