Community code plugin. Review compatibility and verification before install.
Latest release: v1.0.0Download zip
Capabilities
Compatibility
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
The name, description, SKILL.md, index.ts, and the src/* modules all implement a traceback parser, an error database, formatting, three registered tools, and a CLI — exactly what a Python-error diagnosis plugin would need. There are no declared environment variables, required binaries, or config paths that contradict the stated purpose.
Instruction Scope
The SKILL.md and the source instruct the agent to parse tool output (ctx.result) and add annotations via after_tool_call; the code only inspects strings and produces formatted diagnostics. There are no instructions to read unrelated files, environment variables, or to transmit data to external endpoints.
Install Mechanism
There is no install spec (instruction-only skill + source files). That is low risk. However, the repository includes a package-lock.json that lists many large cloud/SDK packages (AWS SDK, Anthropic SDK, Bedrock client, etc.) which are not imported or referenced by the plugin code. This is disproportionate to the plugin's simple functionality and could be an artifact of the developer's environment or an expanded lockfile, but it is unexpected and worth verifying on the published npm package (since package.json declares no runtime dependencies).
Credentials
The skill declares no required environment variables or credentials and the code does not read env vars or config paths. The only proportionality concern is that the package-lock includes SDKs that, if actually installed as dependencies, would normally be associated with services requiring credentials — but package.json does not list them as runtime dependencies, so there is no direct request for credentials in this bundle.
Persistence & Privilege
The plugin does not request persistent elevated privileges. always is false, it registers a normal after_tool_call hook (appropriate for auto-detecting tracebacks), and it does not modify other skills or system-wide agent settings.
Assessment
This skill appears to do what it says: parse Python tracebacks and provide explanations and fixes, and it will auto-annotate tool output when installed. Before installing, consider these checks: 1) verify the published npm package (openclaw-pyfix) metadata on the registry — confirm its package.json (dependencies) matches the source here; 2) inspect the package contents you will install (npm pack / unpack) to ensure the large package-lock contents are not being shipped as runtime dependencies; 3) if you want extra caution, run the plugin locally in an isolated environment and run the test suite (npm install && npm test) to observe behavior; 4) remember the plugin will automatically annotate tool output via after_tool_call, so if you want to restrict automatic annotations you can control plugin activation or permissions in your OpenClaw configuration. The main oddity is a large package-lock.json listing unrelated SDKs — this is likely a leftover dev artifact but worth verifying before trusting the published package.Verification
Tags
PyFix
An OpenClaw plugin for diagnosing and fixing common Python errors.
Created by Riya Verma.
PyFix automatically detects Python tracebacks and provides clear explanations, fix suggestions, and example code to help you resolve errors quickly.
Installation
npm install openclaw-pyfix
Add the plugin to your OpenClaw configuration:
{
"plugins": ["openclaw-pyfix"]
}
Error Categories
| Category | Description | Examples |
|---|---|---|
type | Type-related errors | not subscriptable, unsupported operand |
attribute | Attribute access errors | has no attribute, module has no attr |
import | Import and module errors | ModuleNotFoundError, ImportError |
syntax | Syntax and indentation errors | invalid syntax, unexpected indent |
name | Name resolution errors | NameError, UnboundLocalError |
index | Index out of range errors | list index out of range |
key | Dictionary key errors | KeyError |
value | Value conversion/unpacking errors | invalid literal, too many values to unpack |
io | File and permission errors | FileNotFoundError, PermissionError |
runtime | Runtime errors | ZeroDivisionError, RecursionError |
misc | Other errors | UnicodeDecodeError, JSONDecodeError |
Agent Tools
PyFix registers three tools for the OpenClaw agent:
py_fix- Paste a Python traceback to get diagnosis and fix suggestionspy_explain- Look up a specific Python error type by its codepy_errors- List known Python errors, optionally filtered by category
Auto-Detection
PyFix hooks into after_tool_call to automatically detect Python errors in tool output. When an error matching common Python exception types is found, it adds an annotation with diagnosis and fix suggestions.
Detected error types: TypeError, AttributeError, ImportError, SyntaxError, NameError, IndexError, KeyError, ValueError, FileNotFoundError.
CLI
PyFix adds a pyfix CLI subcommand:
# List all known errors
openclaw pyfix errors
# Filter by category
openclaw pyfix errors type
# Look up a specific error
openclaw pyfix lookup "TypeError: 'int' object is not subscriptable"
# List categories
openclaw pyfix categories
Testing
npm install
npm test
The test suite covers:
- Error database integrity (25+ entries, no duplicates, all fields valid)
- Traceback parsing (full tracebacks, bare error lines, multiple errors)
- Markdown formatting (single results, error lists, multiple results)
License
MIT