Run Orca's coding agents on a VPS
Run Orca's coding agents on a VPS so they keep working when your laptop sleeps: the Orca server on headless Ubuntu, under systemd and off the public internet.
#coding-agents
Filtering by topic #coding-agents · clear
Claude effort levels: max vs high
What Claude's five effort levels really do, which level is your default, and why max drains your Pro, Max or Ultra allowance and can add usage credit charges.
Claude Ultra vs ChatGPT Pro: which $200 tier?
Claude Ultra and ChatGPT Pro both sell a $200 tier. Compare what usage each buys, whether the coding agent is included, and what happens at a limit.
Impeccable: a design skill for coding agents
Impeccable gives coding agents 24 design commands and a detector that flags AI design habits with no model. Install it pinned, then audit a page and check it.
The anti-slop skill for AI coding agents
Anti-slop is a third-party Agent Skill that filters generic AI output from your coding agent. What the three tiers do, and what the core costs per call.
Caveman Skill: Make Coding Agents Say Less
Caveman shortens what a coding agent writes, and its proxy shrinks what the agent reads. Install it pinned to a release, then measure your own sessions.
Nested AGENTS.md files for a monorepo
One root AGENTS.md in a monorepo goes stale and burns context on directories the agent never opens. Here is the nested layout that fixes it.
Cloudflare's security audit skill, sandboxed
Install Cloudflare's security audit skill pinned to a commit, then build the OS-enforced sandbox it needs on a VPS so its leads get tested instead of parked.
Why coding agents ignore your instructions
Your instruction file says stop and the agent does it anyway. The mechanics behind that, and a diagnosis you can run before you rewrite the rule again.
MiMo-Code: Xiaomi's answer to Claude Code
What MiMo-Code actually is: Xiaomi's MIT-licensed terminal coding agent, forked from OpenCode. Read the self-reported benchmark claims before you switch.
Run the Flint coding model on a CPU VPS
Flint has 40B parameters but only 2.5B active per token. Size the RAM from config.json, cap the context, pin the build, and see what runs on a CPU VPS today.
Matt Pocock's skills: which ones to install
Matt Pocock's skills repo ships 27 agent skills. Here is which ones a solo developer on a VPS should install, and which only pay off on a team.
Ponytail: the lazy senior dev agent skill
Ponytail makes a coding agent take the smallest change that works. What it ships, what its own benchmarks show, and how to copy the rule today.
Graft: a codebase map for coding agents
Graft parses your repo with tree-sitter into a codebase map your coding agent queries over MCP, so it stops rediscovering the same structure every session.
Claude Code mods and what they can reach
A Claude Code mod runs TypeScript inside the agent with your credentials in reach. See what it can touch on a VPS and the checks to run before you install one.
Claude Code Mod Examples and Writing Your Own
Three real Claude Code mods read hook by hook, then a small TypeScript mod that refuses VPS lockout commands, checked with plugin validate and plugin test.
AI Agent Examples, and What Each One Needs
Seven AI agent examples people run on a VPS: what starts each one, what it can touch, the model it calls and how much server it needs, from our tested guides.
Autoprompt skill: fewer failures, more tokens
Autoprompt solves 73 of 89 Terminal-Bench 2.1 tasks against a baseline of 60, at roughly 3x the time and 2x the tokens. What that costs on a rented VPS.
dox: keep AGENTS.md current automatically
Your AGENTS.md is wrong three weeks later and the agent trusts it. Use dox to regenerate the file from the repo, then review the diff like code.
Multi-model routing for coding agents
Routing coding agents across models throws away the prompt cache that keeps them cheap. When routing pays, when pinning wins, and the arithmetic behind it.
Paritok token gateway: cheaper agent bills?
Paritok compresses the file reads and tool output your coding agent sends. The project claims 74% fewer tokens. Here is the mechanism and the break-even math.
CLI tools a coding agent needs on a VPS
Install the command line tools a coding agent reaches for on a bare Ubuntu VPS, in three tiers, with the reason each one saves tokens and tool calls.
Prompt injection against coding agents
A coding agent runs on text an attacker can write. Map the real injection surfaces on a server, then rank the defences that actually shrink the damage.
How much RAM does a coding agent VPS need?
One always-on coding agent runs on 4 GB and 2 vCPU. The builds and language servers it launches are what fill the box, and what makes it hang.
AI-assisted code and open source policies
Open source projects disagree on AI-assisted code. Find the project's policy before you open the pull request, and disclose it in the commit trailer.
Coding agent telemetry: what gets sent
Four kinds of traffic leave a coding agent, and only one is unavoidable. Audit any agent from the machine itself, then cut the traffic you did not agree to.
Can Codex use your ChatGPT subscription?
Yes. Codex CLI signs in with your ChatGPT Plus or Pro plan and no API key. Here is what the plan covers, and why the unofficial bridge tools keep breaking.
How to Write Your Own Agent Skill
Write your own agent skill by distilling one real failure: the SKILL.md anatomy, the description line that decides when it fires, and how to test it.