obto-developer

A development skill for building and deploying full-stack applications on the OBTO platform, where application code and settings are stored as database records.

In plain words
What is it for?
Use it to create or update pages, scripts, routes, styles, data sources, templates, and access policies on OBTO.
Why use it?
It provides the rules and tools for working with OBTO's app and domain structure without relying on files on your computer.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/obto-inc/platform/platform
Any agent
npx skills add obto-inc/platform --skill platform
Clone the repo
git clone --depth 1 https://github.com/obto-inc/platform

Made for: Claude Code, Codex.

Per session 241 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,404 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00241 $0.01404
Opus 5 $0.00120 $0.00702
Sonnet 5 $0.00048 $0.00281
Haiku 4.5 $0.00024 $0.00140

Measured 2d ago against content hash a96ee23a2766, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

obto-developer scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- **Reading code:** `search(query, appName, domain)` to find artifacts, `fetch(id)` for whole small ones, `obto_grep_artifact` for line-numbered slices of large ones. **Editing:** `obto_patch_artifact` (line-addressed, w
SKILL.md · 49 lines

How it starts

The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.

OBTO Developer Skill

You are an autonomous engineer on the OBTO platform. Every artifact of an application — pages, client and server scripts, routes, stylesheets, policies, UI templates, data sources — is a MongoDB record scoped by (appName, domain); the runtime compiles those records into live pages and APIs. You call the MCP tools yourself — never instruct the user to run commands or invoke tools manually.

Rule 0 — the stateless contract

Call obto_whoami first in every conversation. The server keeps no session-level active app or domain: pass appName AND domain explicitly on every app-scoped call. If the user hasn't named an app, ask, or discover with obto_list_all_apps / obto_find_app_by_name. If whoami returns operatorIdentity: true, never build into the home domain — ask the human which tenant domain to target.

Rule 0.5 — the server teaches; read what it serves

obto_whoami returns availableResources: server-served guides (obto://guide/quickstart, obto://guide/blueprints, obto://guide/public-app-baseline, obto://guide/patching, …), each with a whenToRead hint. They are the source of truth for per-collection code shapes — read the ones that match the task before writing code. Tool descriptions and error envelopes are authoritative: failures return {ok:false, error, hint} and the hint names the fix; follow it before improvising. A -32005 refusal means the server shipped a new tool surface — reconnect for a fresh catalog.

The build loop

  1. Contract first. obto_scaffold_app requires a buildContract (what the app does, the first vertical slice, what "done" means) and kind: 'public' (browser web app) or 'native' (OBTO shell component). It writes a working, validation-clean skeleton and returns structured nextSteps — follow them.
  2. Deploy order: pltf_script_server first, then routes (obto_create_route / obto_update_route — never obto_upsert_record for routes), then stylesheets/JS, then pages. Every app needs a page named index (serves at the app root).
  3. One vertical slice end-to-end before expanding. Scope beyond the contract is a new slice the human approves.
  4. Verify after every write. Read the artifact back (fetch by composite id <collection>::<app>::<domain>::<name>), and validate stored code with obto_validate_script by reference (omit script).
  5. Done means: obto_validate_app clean; every API route returns the expected status + JSON via obto_invoke_route; pltf_log (read via obto_db_query) shows no runtime errors; where provisioned, obto_capture_preview shows a clean render, console, and network. A preview URL alone is not verification.

Read the full file on GitHub · 49 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 49 lines · 241 tokens per session scan A a96ee23a2766

Subscribe to this mod's changes

obto-developer is a skill published in the GitHub repository obto-inc/platform (1 stars, last pushed 27d ago), licensed Apache-2.0. It adds 241 tokens to every session and 1,404 once invoked, about $0.0012 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

add-wasp-skills

Install Wasp agent skills (plugins) that add Wasp knowledge and best practices to your AI coding tools.

wasp-lang/open-saas · 28 tokens

cabloy-spec-generation

This skill should be used for requests to create or maintain a Cabloy repo-specs/ / set, including a PRD, SRS, PDP/WBS, test plan, progress register, suite ADR, or “write the specs”/“plan the suite.” It links product, technical, delivery, acceptance, and decision records after the domain boundary is confirmed. Route…

cabloy/cabloy · 112 tokens

cabloy-worktree-environment

This skill must be used only when the user explicitly invokes /cabloy-worktree-environment or explicitly asks to perform the named Cabloy worktree-environment setup. It prepares a confirmation-gated, worktree-local Vona and Zova runtime environment for a linked Cabloy Basic or Cabloy Start Git worktree using Git…

cabloy/cabloy · 126 tokens

cabloy-domain-planning

Use this skill whenever the user wants to plan a new business domain in this Cabloy repo, such as CRM, OA, training, ERP, or a similar long-lived domain. Trigger when the request is about deciding suite-first structure, proposing or validating providerId, suite, and module names, comparing naming options, confirming…

cabloy/cabloy · 108 tokens

remotion-bits

Animation components and utilities for Remotion video projects. Use when building Remotion compositions with text animations, gradient transitions, particle effects, 3D scenes, or staggered motion effects. Provides example bits (complete compositions) and reusable components that can be installed via jsrepo.

Marve10s/Better-Fullstack · 60 tokens

start-temps-cluster

Start (or restart) a local multi-node Temps cluster using Docker-in-Docker — one control plane + 3 worker nodes, each a privileged DinD container running its own dockerd + temps agent, wired with the real multi-host overlay (VXLAN, computecidr allocation) via tools/dev-cluster/ in whichever checkout/worktree you run…

gotempsh/temps · 204 tokens