Build an agent
The easiest way to build a production-quality agent.
Assemble a lean main prompt plus proven skills from the registry — every part measured, swappable, and safety-gated.
01
Write skills in markdown
Each behavior is a
SKILL.md — name, triggers, instructions. Fork from the registry or write your own.02
Wire it in two lines
decimalai.init() captures traces and the manifest. Works with LangChain, OpenAI Agents, ADK, or plain Python.03
Route, don’t stuff
router.smart_route(task) hands the agent a ranked shortlist — not your whole folder in the system prompt.04
Eval the bundle
Run your cases with and without the skills, and keep what lifts.
05
Ship it governed
Publish to your workspace — safety-gated, versioned, shared with your team.
support-agent/
├─ skills/
│ ├─ diagnostic-triage-questions.md
│ ├─ billing-dispute-resolution.md
│ └─ sla-breach-response.md
├─ eval/cases.yaml · 21 cases
└─ main.py · the harness
# main.py
import decimalai
decimalai.init(langchain=True)
skills = decimalai.SkillRouter(api_key=DECIMALAI_API_KEY).smart_route(task)
Bundle eval · support-agent
Pass rate without skills43%
Pass rate with bundle76%
Lift+33 pts
illustrative — run it on your own cases with the open eval runner
Playbooks
Pick your agent. Start from its playbook.
Each playbook is a starter kit of proven skills, a pre-tuned router config, and a tutorial — fork the kit, or build it step by step.
Coding agents
8 starter skills, pre-selected
Ship consistent, reviewable output from coding agents.
⌗ silent-failure-review
⌗ unsafe-type-escape-review
Data & SQL agents
8 starter skills, pre-selected
Queries that return what they promise — validated, typed, safe.
⌗ sql-analytics-conventions-mt
⌗ cohort-retention-analysis
⌗ funnel-decomposition
Support agents
8 starter skills, pre-selected
Triage, resolve, and escalate without leaking a thing.
⌗ diagnostic-triage-questions
⌗ billing-dispute-resolution
⌗ sla-breach-response
Security agents
8 starter skills, pre-selected
Guard the agent itself — inputs, outputs, and secrets.
⌗ iam-least-privilege-review
⌗ secret-finding-triage
⌗ threat-model-stride
Research & web agents
8 starter skills, pre-selected
Structured extraction that survives messy pages.
⌗ highlight-extraction
⌗ search-strategy-protocol
⌗ contradiction-map
Ops & infra agents
8 starter skills, pre-selected
Changes to production that never skip the guardrails.
⌗ slo-alert-tuning
⌗ runbook-authoring
⌗ otel-span-naming
Your team’s skills
Author, eval, and share your own — on the same rails.
See skills governance on the registry →Start with the skills for the agent you’re building.
Browse free — no sign-up. Two lines to instrument your agent.
Start the tutorialor browse the registry