finsliparn
Finsliparn is a MCP Server and Claude Code Plugin that adapts Poetiq's ARC-AGI solver methodology
Install / Use
claude mcp add jgabor -- npx -y github:jgabor/finsliparnIf the server publishes to npm under a different name, use that package instead — check the repo README.
MCP Server
Model Context Protocol server
Quality Score
Category
Development & EngineeringSupported Platforms
Tags
Our assessment of finsliparn
finsliparn scores 71/100 on our quality scale, 2435th of 3,356 Development & Engineering skills we index.
Its MCP Server is 2.9 KB long, well organised into 17 sections with 4 code examples: a solid amount of guidance for an agent.
It has 3 GitHub stars, so there is little community track record yet; judge it on its content.
Maintenance, license and trust
- The repository was last updated about 10 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.
- Our last check on 2026-09-18 found the source still online.
- No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
- Its trust signals score 69/100, with 4 cautions from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands.
Automated pattern scan on 2026-09-29. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
finsliparn compared with similar skills
All 4 of these similar skills score higher than finsliparn; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| finsliparn (this skill)by jgabor | 71 | 3 | 10mo ago | MCP Server |
| Agent-Reachby Panniantong | 100 | 86.0k | 13d ago | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 74.0k | today | CLAUDE.md |
| rufloby ruvnet | 100 | 73.4k | today | CLAUDE.md |
| CowAgentby zhayujie | 100 | 47.2k | today | CLAUDE.md |
Frequently asked questions
- How do I install finsliparn?
- Run
claude mcp add jgabor -- npx -y github:jgabor/finsliparn. The install tabs above show the steps for each supported agent. - Which AI agents does finsliparn work with?
- It is written for Claude Code, Claude Desktop and GitHub Copilot, as a MCP Server file. Other agents that read the same format can often use it too.
- Is finsliparn safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. It declares no license and scores 69/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is finsliparn still maintained?
- The repository was last updated about 10 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.
Skill content
View source on GitHubFinsliparn
Swedish: the honer — from finslipa, meaning "to hone", "to fine-tune" or "to put the finishing touches on."
Finsliparn is a Bun-native MCP Server and Claude Code Plugin that transforms single-attempt coding into a multi-attempt, test-validated refinement loop. It adapts the philosophy of Poetiq's ARC-AGI solver to general software development.
🎯 Core Value Proposition
LLMs often generate code that looks correct but fails in edge cases. Finsliparn solves this by enforcing a Refinement Loop:
- Iterate: The LLM implements a solution.
- Validate: Finsliparn runs your actual test suite (not just a syntax check).
- Feedback: Test failures are fed back to the LLM as structured, actionable directives.
- Select: If multiple attempts are made, the best one (highest score, lowest complexity) is selected.
"Test results are the objective truth—not diffs, not prompts, not opinions."
🏗 Architecture
Finsliparn uses a Filesystem-as-IPC architecture to support multiple AI platforms with a single core engine.
graph LR
subgraph "AI Client"
Claude[Claude Code]
Copilot[GitHub Copilot]
end
subgraph "Finsliparn Core"
MCP[MCP Server]
Directive[Directive.md]
Tests[Test Runner]
end
Claude --"Hooks"--> MCP
Copilot --"Agent"--> MCP
MCP --"Writes"--> Directive
Directive --"Reads"--> Claude
Directive --"Reads"--> Copilot
MCP --"Executes"--> Tests
directive.md: The single source of truth. It tells the LLM exactly what to do next (e.g., "Fix failing tests insrc/foo.ts").- Git Worktrees: Every iteration runs in an isolated worktree, ensuring your main branch stays clean until a solution is verified.
🚀 Getting Started
Prerequisites
- Bun v1.3+
- Git
Installation (Development)
# Clone the repository
git clone https://github.com/jgabor/finsliparn.git
cd finsliparn
# Install dependencies
bun install
# Link the plugin (for Claude Code)
# (Coming soon)
📖 Documentation
- Technical Specification (v1.0.0): The core architecture and Claude Code integration.
- Copilot CLI Support (v2.0.0): Adaptation for GitHub Copilot CLI agents.
- Roadmap: Development plan and status.
🛠 Usage
With Claude Code
# Start a refinement session
/finslipa Implement a fibonacci function in src/math.ts
Finsliparn will automatically intercept your edits, run tests, and guide you until the tests pass.
With GitHub Copilot CLI
# Start the agent
copilot run --agent finsliparn "Implement a fibonacci function"
The agent will autonomously loop through the directive.md instructions until completion.
Author
Jonathan Gabor
Related Skills
Agent-Reach
86.0kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
headroom
74.0kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
ruflo
73.4k🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
CowAgent
47.2kOpen-source super AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-agent, multi-model, multi-channel. Lightweight, extensible, one-line install.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
