<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/humans-don-t-install-software-themselves-anymore-their-agents-do-lqxlw7iav" -->

---
title: Humans don't install software themselves anymore, their...
description: AI coding agents are changing how developers onboard to new tools, making manual install scripts and framework-specific quickstarts obsolete. Tigris shipped...
canonical: https://daily.dev/posts/humans-don-t-install-software-themselves-anymore-their-agents-do-lqxlw7iav
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: Humans don't install software themselves anymore, their agents do | daily.dev
og:description: AI coding agents are changing how developers onboard to new tools, making manual install scripts and framework-specific quickstarts obsolete. Tigris shipped...
og:url: https://daily.dev/posts/humans-don-t-install-software-themselves-anymore-their-agents-do-lqxlw7iav
og:image: https://api.daily.dev/og/posts/Lqxlw7IAv.png
og:image:alt: Humans don't install software themselves anymore, their agents do
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Humans don't install software themselves anymore, their agents do

**[Tigris](https://daily.dev/sources/tigris)** · 9 min read · 0 upvotes · 0 comments

## Summary

AI coding agents are changing how developers onboard to new tools, making manual install scripts and framework-specific quickstarts obsolete. Tigris shipped `tigris init --agent`, a CLI command that prints a plain-text setup prompt for any AI agent to execute. The post argues that agents already know a project's stack, conventions, and environment better than any generic setup script can, making them ideal onboarding engineers. Key prompt-writing principles are covered: ask permission before acting, prefer heuristics over mandates, keep secrets out of the context window, and phrase instructions positively. A practical limitation is also addressed — agents can't reliably install their own plugins mid-session, so agent configuration must happen outside an active agent session. The post concludes that agent-native onboarding is becoming the new baseline for developer tooling.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.tigrisdata.com/blog/humans-dont-install-software>

## Questions this post answers

### What does the tigris init --agent command do?

It prints a plain-text setup prompt formatted like a user request, which a coding agent can follow to configure a project with Tigris object storage. Unlike a traditional install script, it makes no assumptions about operating system, package manager, or framework, and delegates decisions like account, environment, and credential storage to the user rather than letting the agent guess.

_daily.dev surfaces practical writeups like this for teams rethinking how CLIs onboard developers via agents._

### How do I refresh Claude Code plugins or skills after installing a new agent plugin?

Use the /reload-skills and /reload-plugins commands in Claude Code to forcibly refresh the skills and plugin lists, since agents don't reliably detect and activate newly installed skills on their own for security reasons (the agent can't tell if new executable code was added legitimately or maliciously). Other agent harnesses like OpenClaw and Hermes are believed to refresh automatically.

_developers wiring plugins into coding agents can track gotchas like this on daily.dev before hitting them in production._

### Why should setup instructions for AI agents use heuristics instead of exact mandates?

Agents already know a project's structure better than a hardcoded script can predict, so giving them a goal and letting them explore (for example, inferring a project name) is more reliable than specifying exact file locations. Exact instructions should be reserved only for things the agent must not guess, such as account, organization, permission scope, and credential storage policy.

_daily.dev helps developers compare emerging prompt-design patterns for agent-driven onboarding and tooling._

## Similar posts on daily.dev

- [From prompt to production: Teams agent setup, simplified](https://daily.dev/posts/from-prompt-to-production-teams-agent-setup-simplified-2hszwfrtv) · DevBlogs · 1 upvotes · 0 comments
- [The IDE Is Losing to the Terminal — Coding Agents Made the Difference](https://daily.dev/posts/the-ide-is-losing-to-the-terminal-coding-agents-made-the-difference-4wqtpmdct) · Medium · 3 upvotes · 0 comments
- [What Nobody's Telling You About AI Agents and Team Development](https://daily.dev/posts/what-nobody-s-telling-you-about-ai-agents-and-team-development-vwl3pashl) · RUBYLAND · 0 upvotes · 0 comments
- [CLI or IDE? Build in verification first](https://daily.dev/posts/cli-or-ide-build-in-verification-first-tv8ffkycw) · The New Stack · 0 upvotes · 0 comments
- [How I Use Agents Without Stopping My Own Growth](https://daily.dev/posts/how-i-use-agents-without-stopping-my-own-growth-ribjpan0v) · Atomic Spin · 62 upvotes · 11 comments

---

Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#prompt-engineering](https://daily.dev/tags/prompt-engineering)

[View this post on daily.dev](https://daily.dev/posts/humans-don-t-install-software-themselves-anymore-their-agents-do-lqxlw7iav)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"Humans don't install software themselves anymore, their agents do","url":"https://daily.dev/posts/humans-don-t-install-software-themselves-anymore-their-agents-do-lqxlw7iav","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/humans-don-t-install-software-themselves-anymore-their-agents-do-lqxlw7iav"},"datePublished":"2026-07-28T14:49:59.953Z","dateModified":"2026-09-14T06:38:59.770Z","description":"AI coding agents are changing how developers onboard to new tools, making manual install scripts and framework-specific quickstarts obsolete. Tigris shipped...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e599a4c1aa1d12554a81ec4a12048827?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e599a4c1aa1d12554a81ec4a12048827?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Tigris","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Tigris","logo":"https://media.daily.dev/image/upload/s--A5ku_6TS--/f_auto/v1716186679/logos/tigris","url":"https://daily.dev/sources/tigris"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/humans-don-t-install-software-themselves-anymore-their-agents-do-lqxlw7iav","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,prompt-engineering","timeRequired":"PT9M"}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Tigris","item":"https://daily.dev/sources/tigris"},{"@type":"ListItem","position":3,"name":"Humans don't install software themselves anymore, their agents do"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/humans-don-t-install-software-themselves-anymore-their-agents-do-lqxlw7iav#faq","mainEntity":[{"@type":"Question","name":"What does the tigris init --agent command do?","acceptedAnswer":{"@type":"Answer","text":"It prints a plain-text setup prompt formatted like a user request, which a coding agent can follow to configure a project with Tigris object storage. Unlike a traditional install script, it makes no assumptions about operating system, package manager, or framework, and delegates decisions like account, environment, and credential storage to the user rather than letting the agent guess. daily.dev surfaces practical writeups like this for teams rethinking how CLIs onboard developers via agents."}},{"@type":"Question","name":"How do I refresh Claude Code plugins or skills after installing a new agent plugin?","acceptedAnswer":{"@type":"Answer","text":"Use the /reload-skills and /reload-plugins commands in Claude Code to forcibly refresh the skills and plugin lists, since agents don't reliably detect and activate newly installed skills on their own for security reasons (the agent can't tell if new executable code was added legitimately or maliciously). Other agent harnesses like OpenClaw and Hermes are believed to refresh automatically. developers wiring plugins into coding agents can track gotchas like this on daily.dev before hitting them in production."}},{"@type":"Question","name":"Why should setup instructions for AI agents use heuristics instead of exact mandates?","acceptedAnswer":{"@type":"Answer","text":"Agents already know a project's structure better than a hardcoded script can predict, so giving them a goal and letting them explore (for example, inferring a project name) is more reliable than specifying exact file locations. Exact instructions should be reserved only for things the agent must not guess, such as account, organization, permission scope, and credential storage policy. daily.dev helps developers compare emerging prompt-design patterns for agent-driven onboarding and tooling."}}]}
```

