A highly disciplined Openclaw Skills extension to automatically verify, organize, and execute fixes for code review feedback.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install receive-feedback
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install receive-feedback using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Receive Feedback?
The receive-feedback capability is a robust, verification-first workflow that processes code reviews, CI/CD linting suggestions, or peer feedback with complete technical rigor. Instead of performatively accepting suggestions, this skill forces the agent to verify every claim against the codebase using concrete tools rather than memory. It completely eliminates common AI limitations like ignoring "out-of-scope" or "pre-existing" issues, ensuring every valid feedback point is resolved with direct code modifications.
As a core part of the Openclaw Skills ecosystem, it allows orchestrators to analyze, verify, and either implement fixes sequentially or spawn parallel subagents for each issue. This ensures that every valid bug receives focused attention and follows a strict Fix-Quality Contract without messy hacks or over-engineered abstractions.
Receive Feedback Use Cases
- Automating the execution of pull request review comments from platforms like GitHub or GitLab.
- Processing and resolving complex linting, security, and compilation feedback generated by CI/CD pipelines.
- Handling cross-session feedback handed off from distinct LLM agents during multi-stage development processes.
- Accelerating pair programming iterations by compiling user suggestions directly into the Openclaw Skills workflow for immediate resolution.
How Receive Feedback Works
- Ingest & Parse: The skill reads the input feedback file, parsing individual feedback items whether they are structured lists or freeform markdown blocks.
- Tool-Based Verification: The orchestrator checks each claim using verification tools (e.g., file readers, test execution commands) to classify them into VALID, INVALID, or NEEDS CLARIFICATION buckets.
- Single-Prompt Confirmation: It outputs a concise summary and proposes to execute all valid fixes at once, asking a single confirmation prompt.
- Fix Quality Dispatching: Upon user confirmation (or subset selection), it applies fixes. If the system supports subagents, they run in parallel; otherwise, the orchestrator applies fixes sequentially.
- Rigorous Refactoring: Each fix is applied strictly adhering to a Fix-Quality Contract, matching regional code architecture while avoiding over-engineering and clutter.
- Reporting & Artifact Generation: Emits a structured table documenting implemented and rejected items with detailed citations.
Receive Feedback Setup
To run this workflow, call the execution script from your CLI workspace. Ensure your workspace is set up and configured with the required subagent configurations or fallback sequential engines.
To run the skill against a feedback file:
receive-feedback reviews/pr-123-feedback.md
Make sure the following configuration files are present in your workspace root:
VERIFICATION.mdfor defining validation scriptsEVALUATION.mdto customize classification boundariesRESPONSE.mdto format the final execution table
Receive Feedback Data Schema & Taxonomy
The workspace directories utilize structured markdown schemas to pass data through the Gates:
| File | Purpose | Key Elements |
|---|---|---|
VERIFICATION.md |
Logic guidelines for testing code correctness | Tool parameters, file verification paths, test scripts |
EVALUATION.md |
Guidelines for assigning item states | Criteria for VALID, INVALID, and NEEDS CLARIFICATION |
RESPONSE.md |
Format for output artifact generation | Implemented table, Rejected table, User-excluded listing |
Feedback Input Format
Accepts numbered or bulleted markdown lists:
1. Remove unused import on line 15
2. Add error handling to the API call
Receive Feedback Advanced Features
- Fix-Quality Contract Enforcement: Ensures idiomatic refactoring by forcing strict compliance to surrounding modules, avoiding inline hacks or lazy fixes.
- Parallel Subagent Orchestration: Automatically scales up parallel subagents when supported, processing multiple distinct fixes concurrently while isolating execution contexts.
- Single Batch Gatekeeping: Minimizes developer friction by packing multiple actions into a single-point confirmation prompt instead of annoying multi-turn conversations.
- Rigorous Rejection Audits: Requires concrete, tooling-derived proof to reject a feedback item, raising the quality bar for collaborative AI environments powered by Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Release Notes Generator
An automated AI agent skill that analyzes Git histories, extracts merged pull requests, and generates strict Keep a Changelog compliant documentation.
Review Plan
An automated AI agent skill that executes a rigorous, five-lens implementation plan review before executing code modifications.
Prompt Improver
An AI-agent tool designed to automatically audit and transform vague engineering prompts into clear, highly optimized instructions for code implementation, debugging, refactoring, and testing.
Remix v2 Data Flow Review
An advanced automated code review skill for analyzing and optimizing Remix v2 loader, action, and revalidation data flows.
opc-technical-due-diligence
A specialized technical due diligence screening tool designed for pre-investment feasibility analysis, background verification, and patent validation.
LLM Artifacts Detection
An automated detection framework designed to scan codebases and eliminate code quality issues, dead code, over-abstraction, and verbose style patterns introduced by AI coding agents.
Featured*
