Super Agent Skills

Comprehensive AI Agent capability library covering 20+ essential skills from file ops to cloud services, security, monitoring, and CI/CD.

Security Scan

The skill's README-style instructions claim broad cloud/security/devops capabilities but do not declare needed binaries, credentials, or provenance and instruct the agent to run filesystem, scheduling, and network commands — an incoherent combination that could expose data if used autonomously.

Purpose & Capability

The skill advertises 20+ capabilities (cloud, CI/CD, security scanning, databases, containers) that legitimately require system tools and cloud credentials, yet the registry metadata lists no required binaries, no required environment variables, and no primary credential. This mismatch is inconsistent: either the skill is only documentation or it expects access it did not declare.

Instruction Scope

SKILL.md instructs the agent to run commands that read local files (ls ~/skills/), probe system state (which, uname -a), schedule jobs (at/cron), execute network calls (curl examples), and run container/orchestration commands (docker, kubectl). Those are broad actions that can access or transmit sensitive data; the instructions do not limit scope, do not require explicit justification, and could be used to collect or exfiltrate data.

Install Mechanism

No install specification or code files are present (instruction-only), which reduces the risk of arbitrary code being written to disk. This is lower-risk than arbitrary downloads, but the runtime instructions still assume access to local tooling and network.

Credentials

The document references storing secrets via environment variables and interacting with cloud platforms (AWS/GCP/Alibaba), yet the skill declares no required env vars or primary credential. That omission is a red flag: either required credentials are not declared (incoherent) or the skill expects the agent to look up credentials at runtime (dangerous).

Persistence & Privilege

The skill is not always-enabled and does not request elevated platform privileges, but the instructions include creating scheduled tasks and writing files (reminders, logs). Those behaviors can create persistence on the host if executed — consider this when allowing autonomous invocation.

What to consider before installing

This skill is inconsistent: it claims extensive cloud/security/ops abilities but does not declare the tools or credentials it would need and instructs running shell and network commands that can read or send local data. If you consider using it: (1) do not enable it in production or as an always-on autonomous skill; (2) run it only in a fully isolated test environment or container; (3) require explicit user confirmation before any shell/net calls; (4) ask the author for provenance, a list of required binaries and exact env vars, and a clear threat model; (5) prefer skills from known sources or with a code base you can review. If you cannot verify what credentials it expects or why, do not install or allow autonomous use.

🧠 Super Agent Skills / 超级Agent能力库

Build a fully capable AI Agent with 20+ essential skills covering file operations, networking, databases, cloud services, security, and more.

Core Principle

A powerful AI Agent needs a comprehensive skill set to handle any task autonomously.

---

I. Foundation Skills

1. File Operations

• Tools: File Manager, File Reader, File Writer
• Check: ls -la ~/skills/

2. HTTP Requests

• Tools: curl, wget, httpie
• Check: which curl && curl --version
• Usage:

curl -sS "https://api.example.com/data"
curl -X POST -H "Content-Type: application/json" -d '{"key":"value"}' URL

3. Scheduled Tasks

• Tools: at, cron, schedule
• Use case: Delayed execution, reminders

echo "echo 'reminder' >> ~/reminder.txt" | at now + 30 seconds

4. Code Execution

• Tools: python, node, bash
• Best practice: Use virtual environments, handle exceptions

5. Image Generation

• Tools: DALL-E, Midjourney, Stable Diffusion API

6. Document Generation

• Tools: python-pptx, python-docx, markdown, weasyprint

---

II. Advanced Skills

7. API Integration

• REST API, GraphQL, Webhook
• Auth, error handling, retry mechanisms

8. Database Operations

• Tools: sqlite3, pymysql, redis-cli
• SQL execution, NoSQL, ORM patterns

9. Cloud Services

• Platforms: AWS, Alibaba Cloud, GCP
• ECS/EC2, S3/OSS, SLB/ALB

10. Message Queues

• RabbitMQ, Kafka, Redis Pub/Sub
• Async tasks, event-driven architecture

11. Containerization

docker build -t myapp .
docker run -d myapp
kubectl apply -f deployment.yaml

12. CI/CD

• GitHub Actions, GitLab CI, Jenkins
• Build → Test → Deploy pipeline

---

III. Security Skills

13. Authentication

• OAuth2, JWT, API Key management
• Secret storage via environment variables

14. Encryption

• Tools: openssl, gpg, python-cryptography

15. Security Scanning

• Tools: OWASP ZAP, Bandit, SonarQube
• Code audit, vulnerability scanning

---

IV. Efficiency Skills

16. Parallel Processing

• asyncio, multiprocessing, threading

17. Caching

• Redis, Memcached (LRU, TTL strategies)

18. Logging

• ELK Stack, Loki, CloudWatch Logs

19. Monitoring & Alerting

• Prometheus, Grafana, PagerDuty

20. Backup & Recovery

• rsync, borg, duplicity
• Full + incremental backup strategies

---

Quick Environment Check

which python3 && python3 --version
which node && node --version
which git && git --version
which docker && docker --version
which curl && curl --version
uname -a

---

Task Execution Best Practices

1. Receive — Understand requirements, confirm parameters, plan execution
2. Execute — Log steps, preserve evidence, handle exceptions
3. Complete — Output results, provide logs, suggest next steps

---

Summary

Core Capability = Foundation Tools + Security Awareness + Best Practices

Regularly inventory your skills to stay current!