← Back to Skills Marketplace
xinywa Coding Helper
by
alice_wang
· GitHub ↗
· v1.0.2
· MIT-0
123
Downloads
1
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install xinywacodinghelper
Description
Safely implement minimal, well-tested, and local-pattern-consistent code changes in existing codebases with clear understanding, logging, and resumable progr...
Usage Guidance
This skill appears to do what it says (safe, resumable changes in complex repos) and requests no credentials, but it assumes it can read/write the repository and persistent host paths (/home/vibe/aksclaw-data/...). Before installing, confirm: (1) the agent environment allows and restricts filesystem writes to those paths (or update the skill to use an appropriate configured directory); (2) the routing rule that it only runs for COMPLEX_CODE_TASK is enforced in your agent/router; (3) stored indices/logs don’t leak sensitive code to shared storage or backups. If you allow autonomous invocation, consider restricting this skill’s use to trusted agents or require manual approval for runs that touch production repositories.
Capability Analysis
Type: OpenClaw Skill
Name: xinywacodinghelper
Version: 1.0.2
The skill bundle provides a structured framework for an AI agent to perform code refactoring and maintenance tasks. It includes detailed instructions in SKILL.md and WATCHER.md for progress tracking, error handling, and repository indexing using local paths (e.g., /home/vibe/aksclaw-data/). No malicious code, data exfiltration attempts, or harmful prompt injections were identified; the instructions are focused on ensuring visibility and safety during complex coding tasks.
Capability Assessment
Purpose & Capability
Name/description (safe, minimal code changes in large repos) matches the SKILL.md: it requires repo access, incremental scanning, indexing, resumable runs, and visible progress. There are no unrelated env vars, binaries, or third-party credentials requested.
Instruction Scope
Instructions stay within the stated purpose (understand, plan, implement, validate, review) and explicitly forbid silent runs and blind repo scans. They do, however, require reading and writing repository files and maintaining progress/index files; this is expected for the purpose but broad in scope (full read/write access to the repo and to host paths).
Install Mechanism
Instruction-only skill with no install spec and no code files — lowest install risk. Nothing is downloaded or executed by an installer.
Credentials
The skill declares no env vars or credentials (appropriate). However, the SKILL.md references hard-coded host filesystem locations (/home/vibe/aksclaw-data/index/ and /home/vibe/aksclaw-data/runs/) which are not declared as required config paths. That implicit requirement for writable persistent storage should be validated.
Persistence & Privilege
Skill will create and update persistent indices and run logs under the referenced host directories; always:false and no cross-skill config modifications. This persistent storage increases blast-radius of a compromised agent account if those paths are shared or accessible by others.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install xinywacodinghelper - After installation, invoke the skill by name or use
/xinywacodinghelper - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
Summary: This version adds detailed operational documentation and machine-optimized process requirements for safe, minimal code changes in large repositories.
- Added RECOVERY.md, ROUTING.md, WATCHER.md, and templates/progress_template.md with supporting operational and process details.
- SKILL.md overhauled to enforce mandatory repo access checks, progressive understanding, and machine-optimized index usage.
- All execution is now resumable, logged per-step, and prohibits silent operation for traceability.
- Confidence scoring and structured error/fail-fast handling are now required.
- Output and log directories specified for consistent run tracking and post-run cleanup.
- Skill now restricted to COMPLEX_CODE_TASK routing only; simple tasks are excluded.
v1.0.1
- Added a clear skill routing section to classify requests as code or non-code tasks before responding.
- Defined explicit criteria for what constitutes a CODE TASK and a NON-CODE TASK.
- All existing brownfield code change principles and processes remain unchanged.
- No code or logic changes — documentation update only.
v1.0.0
init
Metadata
Frequently Asked Questions
What is xinywa Coding Helper?
Safely implement minimal, well-tested, and local-pattern-consistent code changes in existing codebases with clear understanding, logging, and resumable progr... It is an AI Agent Skill for Claude Code / OpenClaw, with 123 downloads so far.
How do I install xinywa Coding Helper?
Run "/install xinywacodinghelper" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is xinywa Coding Helper free?
Yes, xinywa Coding Helper is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does xinywa Coding Helper support?
xinywa Coding Helper is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created xinywa Coding Helper?
It is built and maintained by alice_wang (@xinyue-wang); the current version is v1.0.2.
More Skills