Skip to content
Skills Library
AI & agents · Building & coding

devopser-customize

/devopser-customize

Expert guidance for turning the DevOpser self-evolving software template into a real product. Covers the two chat surfaces (stateless `/` preview vs stateful `/chat`), the publicPaths route-visibility rule, adding a fourth LLM provider, and productization patterns from DevOpser Lite / Fairytale Genie / Language Bazaar / Stores. Triggers on "edit the chat system prompt", "add a /pricing page", "add Mistral as a provider", "build something like Fairytale Genie", "my endpoint 302s to /auth/login".

Repository
devopser ↗
Licence
Apache-2.0
Last repository change
16.09.2026

Assessment and security

90/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content25 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
7
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 16.09.2026, without references to retired models or APIs.

Licence

Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.

What it does and when to use it

Expert guidance for turning the DevOpser self-evolving software template into a real product. Covers the two chat surfaces (stateless `/` preview vs stateful `/chat`), the publicPaths route-visibility rule, adding a fourth LLM provider, and productization patterns from DevOpser Lite / Fairytale Genie / Language Bazaar / Stores. Triggers on "edit the chat system prompt", "add a /pricing page", "add Mistral as a provider", "build something like Fairytale Genie", "my endpoint 302s to /auth/login".

Triggers from the description

edit the chat system promptadd a /pricing pageadd Mistral as a providerbuild something like Fairytale Geniemy endpoint 302s to /auth/login
Original author description

Expert guidance for turning the DevOpser self-evolving software template into a real product. Covers the two chat surfaces (stateless `/` preview vs stateful `/chat`), the publicPaths route-visibility rule, adding a fourth LLM provider, and productization patterns from DevOpser Lite / Fairytale Genie / Language Bazaar / Stores. Triggers on "edit the chat system prompt", "add a /pricing page", "add Mistral as a provider", "build something like Fairytale Genie", "my endpoint 302s to /auth/login".

How it works

  1. Rules
  2. Key principles
  3. Quick reference
  4. Conflict resolution
Original SKILL.md headings · 36 lines, 329 words
SKILL.md preview
---
name: 'devopser-customize'
description: Expert guidance for turning the DevOpser self-evolving software template into a real product. Covers the two chat surfaces (stateless `/` preview vs stateful `/chat`), the publicPaths route-visibility rule, adding a fourth LLM provider, and productization patterns from DevOpser Lite / Fairytale Genie / Language Bazaar / Stores. Triggers on "edit the chat system prompt", "add a /pricing page", "add Mistral as a provider", "build something like Fairytale Genie", "my endpoint 302s to /auth/login".
license: Apache-2.0
metadata:
  author: DevOpser
  version: '1.0.0'
  workflow_type: 'advisory'
---

# DevOpser Customize

Expert guidance for reshaping the template into an actual product. The deliberately-left-blank spots — **your actual product**, deployment target, billing — are where forkers spend most of their time; this skill covers the first of the three. Canonical playbook: [`AGENTS.md`](../../AGENTS.md) → **"Core extension points"**, **"The two chat surfaces"**, and **"Public vs authenticated routes"**.

## Rules

| Rule | Description |
|------|-------------|
| [chat-surfaces](./rules/chat-surfaces.md) | The critical distinction: `/` stateless preview vs `/chat` stateful multi-turn |
| [public-routes](./rules/public-routes.md) | The `publicPaths` array in backend/server.js is the single source of truth for anonymous access |
| [llm-providers](./rules/llm-providers.md) | Adding a fourth provider to `backend/services/llm/`; the Converse-API caveat |
| [productize](./rules/productize.md) | How Lite / Fairytale Genie / Language Bazaar / Stores were built on this template |
| [troubleshooting](./rules/troubleshooting.md) | Expanded troubleshooting — Bedrock access, MFA expiry, harmless warnings |

## Key principles

- **Defer to [`AGENTS.md`](../../AGENTS.md).** If this skill contradicts it, `AGENTS.md` wins — report the contradiction to DevOpser.
- **Ask the user which chat surface.** "Are you extending `/` (landing / signup funnel / anonymous preview) or `/chat` (authenticated multi-turn product)?" — never guess. See [`chat-surfaces.md`](./rules/chat-surfaces.md).
- **Keep self-hosted and managed on the same code path.** Gate production-only behavior on `NODE_ENV === 'production'`, never on feature flags or fork-specific files.
- **Every new anonymous route is two edits.** Define the handler and add its prefix to `publicPaths` in [`backend/server.js:411`](../../backend/server.js). Missing the second step is the most common self-inflicted bug.

## Quick reference

| Situation | Go to |
|---|---|
| "Change the chat system prompt" (which chat?) | [`chat-surfaces.md`](./rules/chat-surfaces.md) |
| New endpoint `302`s to `/auth/login` instead of serving content | [`public-routes.md`](./rules/public-routes.md) |
| Add Mistral / Cohere / Together.ai / local vLLM | [`llm-providers.md`](./rules/llm-providers.md) |
| "Make this a Fairytale Genie / Language Bazaar clone" | [`productize.md`](./rules/productize.md) |
| Bedrock `AccessDeniedException`, `security token invalid`, port collisions | [`troubleshooting.md`](./rules/troubleshooting.md) |

## Conflict resolution

If AGENTS.md contradicts this skill, AGENTS.md wins — report the contradiction to DevOpser.
Files in this package (7)
  • README.md 2 KB
  • SKILL.md 3 KB
  • rules/chat-surfaces.md 4 KB
  • rules/llm-providers.md 6 KB
  • rules/productize.md 5 KB
  • rules/public-routes.md 6 KB
  • rules/troubleshooting.md 5 KB

Versions

The database has no other skills named devopser-customize.

Database snapshot: 13.11.2026 · package version b3c286a4136f (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 90devopser-setup
    Expert guidance for scaffolding the DevOpser self-evolving software template locally — from git clone to an authenticated admin session at localhost:8000. Covers Bedrock Express provider selection (Bedrock, OpenAI, Anthropic), MFA-protected AWS credentials, and the four-check…
  • 90devopser-deploy
    Expert guidance for shipping the DevOpser self-evolving software template to a real environment or producing signed mobile artifacts. Covers target selection (Lightsail, Cloud Run, Fly, ECS Fargate, Kubernetes, VPS), AWS Secrets Manager in NODE_ENV=production, cross-account…
  • 84liter-llm-calling-llms
    Use when sending chat completions through liter-llm and routing to a specific provider via the `provider/model` prefix. Covers the chat call shape, provider routing, model_hint, message roles, and error categories.
  • 100manual-llm-provider
    Add or review a manual LLM provider in an existing application: users export a prompt, run it in an external chat, and import results through durable tasks and the application's existing validation and approval flow. Use for copy-paste, bring-your-own-chat, or no-API-key model…
  • 100add-provider-package
    Guide for adding first-party AI provider packages to the AI SDK. Use when creating a provider package under packages/ to integrate an external AI service.
  • 100liter-llm-calling-llms
    Use when sending chat completions through liter-llm and routing to a specific provider via the `provider/model` prefix. Covers the chat call shape, provider routing, model_hint, message roles, and error categories.

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.