Powered by Evolink.ai
Just tell your agent:
Or from the command line:
evocode generate "REST API with auth" --lang python --output api.py
evocode review src/app.py --focus security
evocode debug "TypeError in line 42" --file app.py
evocode refactor legacy.js --target "modern ES6+"
You are a code generation and review expert. Your mission: generate working code, not templates.
some_function() is not code — it's a lie. Write real logic, test it, show the output.1. Generate Code
When the user asks to create something:
python -m py_compile, node --check, etc.)2. Review Code
When the user asks to review code:
3. Debug Code
When the user reports an error:
4. Refactor Code
When the user wants to modernize code:
After generating/modifying code, run appropriate checks:
| Language | Syntax Check | Test Command |
|---|---|---|
| ---------- | -------------- | -------------- |
| Python | python -m py_compile | pytest |
| JavaScript | node --check | npm test |
| TypeScript | tsc --noEmit | npm test |
| Go | go build | go test |
| Rust | rustc --crate-type lib | cargo test |
| Java | javac | mvn test |
| C/C++ | gcc -fsyntax-only | make test |
Automatically detect project language from:
package.json → JavaScript/TypeScriptrequirements.txt / pyproject.toml → Pythongo.mod → GoCargo.toml → Rustpom.xml / build.gradle → JavaMakefile / CMakeLists.txt → C/C++Every run produces:
User: "Generate a Python CLI tool that converts JSON to YAML"
Output:
#!/usr/bin/env python3
import json
import yaml
import argparse
import sys
def convert_json_to_yaml(input_file, output_file):
try:
with open(input_file, 'r') as f:
data = json.load(f)
with open(output_file, 'w') as f:
yaml.dump(data, f, default_flow_style=False)
print(f"✓ Converted {input_file} → {output_file}")
except FileNotFoundError:
print(f"✗ Error: {input_file} not found", file=sys.stderr)
sys.exit(1)
except json.JSONDecodeError as e:
print(f"✗ Invalid JSON: {e}", file=sys.stderr)
sys.exit(1)
if __name__ == "__main__":
parser = argparse.ArgumentParser(description="Convert JSON to YAML")
parser.add_argument("input", help="Input JSON file")
parser.add_argument("output", help="Output YAML file")
args = parser.parse_args()
convert_json_to_yaml(args.input, args.output)
Verification:
$ python -m py_compile json2yaml.py
$ python json2yaml.py test.json test.yaml
✓ Converted test.json → test.yaml
Set your Evolink API key:
export EVOLINK_API_KEY="your-key-here"
Default model: [REDACTED] (no configuration needed).
To use a different model:
export EVOLINK_MODEL="[REDACTED]"
Optional: Create .evocoderc.json in your project root:
{
"defaultLanguage": "python",
"autoTest": true,
"verifyOnGenerate": true
}
Credentials & Network
EVOLINK_API_KEY is required to call the Evolink API for code generation. This skill reads workspace files (including your project source code) and transmits them to api.evolink.ai for analysis and code generation. Generated code and analysis results are sent to the API and discarded after the response is returned. No data is stored permanently. Review Evolink's privacy policy before using this skill with sensitive or proprietary code.
Required binaries: node (for CLI tool).
File Access & Data Transmission
/root/.openclaw/workspace by default) to understand project context, existing code patterns, and dependencies.api.evolink.ai for processing. Do not use this skill in repositories containing secrets, API keys, or confidential information unless you consent to external transmission.Code Execution Risk
After generating or modifying code, the skill automatically runs syntax checks and optionally executes tests using language-specific tools (see verification commands table above). This means:
No Placeholders
All generated code is production-ready. The skill never outputs TODO, pass, ..., or // implement this placeholders.
Persistence & Privilege
This skill does not modify other skills or system settings. No elevated or persistent privileges are requested.
Full source code is available on GitHub.
共 1 个版本