Add folder anatomy (scripts/agent.py + references/api-reference.md) for 648 cybersecurity skills

Complete skill folder anatomy across all cybersecurity skills:
- scripts/agent.py: 80-150 line Python agents using real libraries (impacket,
  boto3, azure-mgmt-*, kubernetes, pefile, yara, scapy, shodan, stix2, etc.)
- references/api-reference.md: real API documentation with method signatures
- LICENSE: MIT license for all skill folders
This commit is contained in:
mukul975
2026-03-10 21:02:12 +01:00
parent c74d52fa30
commit 27c6414ca5
1390 changed files with 106806 additions and 0 deletions
@@ -0,0 +1,21 @@
MIT License
Copyright (c) 2025 Anthropic Agent Skills Contributors
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
@@ -0,0 +1,45 @@
---
name: implementing-cloud-workload-protection
description: >
Implements cloud workload protection using boto3 and google-cloud APIs for runtime
security monitoring, process anomaly detection, and file integrity checking on EC2/GCE
instances. Scans for cryptomining, reverse shells, and unauthorized binaries.
Use when building runtime security controls for cloud compute workloads.
---
# Implementing Cloud Workload Protection
## Instructions
Monitor cloud workloads for runtime threats by checking process lists, network
connections, file integrity, and resource utilization anomalies.
```python
import boto3
ssm = boto3.client("ssm")
# Run command on EC2 instances to check for suspicious processes
response = ssm.send_command(
InstanceIds=["i-1234567890abcdef0"],
DocumentName="AWS-RunShellScript",
Parameters={"commands": ["ps aux | grep -E 'xmrig|minerd|cryptonight'"]},
)
```
Key protection areas:
1. Process monitoring for cryptominers and reverse shells
2. File integrity monitoring on critical system files
3. Network connection auditing for C2 callbacks
4. Resource utilization anomaly detection (CPU spikes)
5. Unauthorized binary detection via hash comparison
## Examples
```python
# Check for unauthorized outbound connections
ssm.send_command(
InstanceIds=instances,
DocumentName="AWS-RunShellScript",
Parameters={"commands": ["ss -tlnp | grep ESTABLISHED"]},
)
```
@@ -0,0 +1,61 @@
# API Reference: Implementing Cloud Workload Protection
## AWS SSM Run Command (boto3)
```python
import boto3
ssm = boto3.client("ssm")
# Execute command on instances
resp = ssm.send_command(
InstanceIds=["i-abc123"],
DocumentName="AWS-RunShellScript",
Parameters={"commands": ["ps aux"]},
TimeoutSeconds=60,
)
command_id = resp["Command"]["CommandId"]
# Get output
output = ssm.get_command_invocation(
CommandId=command_id, InstanceId="i-abc123"
)
print(output["StandardOutputContent"])
```
## CloudWatch CPU Monitoring
```python
cw = boto3.client("cloudwatch")
resp = cw.get_metric_statistics(
Namespace="AWS/EC2", MetricName="CPUUtilization",
Dimensions=[{"Name": "InstanceId", "Value": "i-abc123"}],
StartTime=start, EndTime=end, Period=300,
Statistics=["Average"],
)
```
## Key Detection Commands
| Threat | Command |
|--------|---------|
| Cryptominer | `ps aux \| grep -iE 'xmrig\|minerd'` |
| Reverse shell | `ss -tlnp \| grep ESTAB` |
| File integrity | `rpm -Va \| grep '^..5'` |
| Unauthorized binaries | `find /tmp -executable -type f` |
| Cron persistence | `crontab -l; ls /etc/cron.d/` |
## GuardDuty Integration
```python
gd = boto3.client("guardduty")
findings = gd.list_findings(DetectorId="detector-id")
for fid in findings["FindingIds"]:
detail = gd.get_findings(DetectorId="detector-id", FindingIds=[fid])
print(detail["Findings"][0]["Type"])
```
### References
- SSM Run Command: https://docs.aws.amazon.com/systems-manager/latest/userguide/run-command.html
- CloudWatch: https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/cloudwatch.html
- GuardDuty: https://docs.aws.amazon.com/guardduty/latest/ug/
@@ -0,0 +1,186 @@
#!/usr/bin/env python3
"""Agent for cloud workload protection on AWS EC2 instances."""
import os
import json
import time
import argparse
from datetime import datetime
import boto3
from botocore.exceptions import ClientError
def get_running_instances(session, filters=None):
"""List all running EC2 instances."""
ec2 = session.client("ec2")
params = {"Filters": [{"Name": "instance-state-name", "Values": ["running"]}]}
if filters:
params["Filters"].extend(filters)
instances = []
paginator = ec2.get_paginator("describe_instances")
for page in paginator.paginate(**params):
for res in page["Reservations"]:
for inst in res["Instances"]:
instances.append({
"instance_id": inst["InstanceId"],
"type": inst["InstanceType"],
"ip": inst.get("PrivateIpAddress", ""),
"launch_time": str(inst["LaunchTime"]),
})
return instances
def run_ssm_command(session, instance_ids, commands):
"""Execute commands on instances via SSM Run Command."""
ssm = session.client("ssm")
resp = ssm.send_command(
InstanceIds=instance_ids,
DocumentName="AWS-RunShellScript",
Parameters={"commands": commands},
TimeoutSeconds=60,
)
command_id = resp["Command"]["CommandId"]
time.sleep(5)
results = {}
for iid in instance_ids:
try:
output = ssm.get_command_invocation(CommandId=command_id, InstanceId=iid)
results[iid] = {
"status": output["Status"],
"stdout": output.get("StandardOutputContent", ""),
"stderr": output.get("StandardErrorContent", ""),
}
except ClientError as e:
results[iid] = {"status": "Error", "error": str(e)}
return results
def scan_for_cryptominers(session, instance_ids):
"""Detect cryptomining processes on instances."""
commands = [
"ps aux | grep -iE 'xmrig|minerd|cryptonight|stratum|nicehash' | grep -v grep",
"find /tmp /var/tmp /dev/shm -name '*.sh' -o -name 'config.json' 2>/dev/null | head -20",
]
results = run_ssm_command(session, instance_ids, commands)
findings = []
for iid, result in results.items():
if result.get("stdout", "").strip():
findings.append({
"instance_id": iid,
"type": "cryptominer",
"severity": "CRITICAL",
"output": result["stdout"].strip(),
})
return findings
def scan_for_reverse_shells(session, instance_ids):
"""Detect potential reverse shell connections."""
commands = [
"ss -tlnp 2>/dev/null | grep ESTAB | grep -vE ':443|:80|:22|:8089'",
"ls -la /dev/tcp 2>/dev/null; ls -la /proc/*/fd 2>/dev/null | grep socket | head -20",
]
results = run_ssm_command(session, instance_ids, commands)
findings = []
for iid, result in results.items():
if result.get("stdout", "").strip():
findings.append({
"instance_id": iid,
"type": "suspicious_connections",
"severity": "HIGH",
"output": result["stdout"].strip(),
})
return findings
def check_file_integrity(session, instance_ids):
"""Check integrity of critical system files."""
commands = [
"rpm -Va 2>/dev/null | grep -E '^..5' | head -20 || "
"debsums -c 2>/dev/null | head -20",
"find /usr/bin /usr/sbin -newer /var/log/lastlog -type f 2>/dev/null | head -20",
]
results = run_ssm_command(session, instance_ids, commands)
findings = []
for iid, result in results.items():
if result.get("stdout", "").strip():
findings.append({
"instance_id": iid,
"type": "file_integrity",
"severity": "MEDIUM",
"modified_files": result["stdout"].strip().splitlines(),
})
return findings
def check_cpu_anomaly(session, instance_ids):
"""Detect CPU usage anomalies via CloudWatch."""
cw = session.client("cloudwatch")
anomalies = []
for iid in instance_ids:
resp = cw.get_metric_statistics(
Namespace="AWS/EC2",
MetricName="CPUUtilization",
Dimensions=[{"Name": "InstanceId", "Value": iid}],
StartTime=datetime.utcnow().replace(hour=0, minute=0),
EndTime=datetime.utcnow(),
Period=300,
Statistics=["Average"],
)
for dp in resp.get("Datapoints", []):
if dp["Average"] > 90:
anomalies.append({
"instance_id": iid,
"cpu_avg": round(dp["Average"], 1),
"timestamp": str(dp["Timestamp"]),
"severity": "HIGH",
})
return anomalies
def main():
parser = argparse.ArgumentParser(description="Cloud Workload Protection Agent")
parser.add_argument("--profile", default=os.getenv("AWS_PROFILE"))
parser.add_argument("--region", default=os.getenv("AWS_DEFAULT_REGION", "us-east-1"))
parser.add_argument("--output", default="cwp_report.json")
parser.add_argument("--action", choices=[
"list", "cryptominer", "reverse_shell", "integrity", "cpu", "full_scan"
], default="full_scan")
args = parser.parse_args()
session = boto3.Session(profile_name=args.profile, region_name=args.region)
report = {"generated_at": datetime.utcnow().isoformat(), "findings": {}}
instances = get_running_instances(session)
instance_ids = [i["instance_id"] for i in instances]
report["instances"] = instances
print(f"[+] Running instances: {len(instances)}")
if args.action in ("cryptominer", "full_scan") and instance_ids:
findings = scan_for_cryptominers(session, instance_ids)
report["findings"]["cryptominers"] = findings
print(f"[+] Cryptominer detections: {len(findings)}")
if args.action in ("reverse_shell", "full_scan") and instance_ids:
findings = scan_for_reverse_shells(session, instance_ids)
report["findings"]["reverse_shells"] = findings
print(f"[+] Suspicious connections: {len(findings)}")
if args.action in ("integrity", "full_scan") and instance_ids:
findings = check_file_integrity(session, instance_ids)
report["findings"]["file_integrity"] = findings
print(f"[+] File integrity issues: {len(findings)}")
if args.action in ("cpu", "full_scan") and instance_ids:
anomalies = check_cpu_anomaly(session, instance_ids)
report["findings"]["cpu_anomalies"] = anomalies
print(f"[+] CPU anomalies: {len(anomalies)}")
with open(args.output, "w") as f:
json.dump(report, f, indent=2, default=str)
print(f"[+] Report saved to {args.output}")
if __name__ == "__main__":
main()