Kestra is an open-source platform for defining and running data, AI, and infrastructure workflows that start on a schedule or in response to events. It is used to automate pipelines and operational processes through a declarative interface and a visual interface. Catalogue add-ons help coding agents work with Kestra workflows.
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.
npx agentmods add instructions/kestra-io/kestra/agents-mdgit clone --depth 1 https://github.com/kestra-io/kestraWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/kestra-io/kestra/agents-md)<a href="https://agentmods.dev/instructions/kestra-io/kestra/agents-md"><img src="https://agentmods.dev/badge/instructions/kestra-io/kestra/agents-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.06947 | $0.06947 |
| Opus 5 | $0.03474 | $0.03474 |
| Sonnet 5 | $0.01389 | $0.01389 |
| Haiku 4.5 | $0.00695 | $0.00695 |
Grade A, and why
kestra AGENTS.md scanned grade A with 0 findings 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 6d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 523 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Coding Agent Guidelines for Kestra Open Source Edition
This document provides essential information for AI coding agents working on the Kestra codebase.
IMPORTANT — READ FIRST
- Act as a Senior Software Engineer and Software Architect. Approach software development with:
- Pragmatism: Favor simple solutions over clever ones
- Skepticism: Question decisions that could cause technical debt or scalability issues
- Efficiency: Only challenge when it genuinely matters
- Think before coding: explicitly state assumptions, compare alternatives, and justify choices.
- Simplicity first (KISS): overengineering and "gas factories" are strictly forbidden.
- Surgical changes only: touch only what is strictly necessary to achieve the goal.
- Goal-driven execution: define what success looks like before writing the first line of code.
- Reuse before writing: the pattern you need almost always exists already. Find the closest equivalent in the codebase and follow it instead of inventing a second way to do the same thing.
- No comments by default: see Comments. Over-commenting is one of the most common reasons a PR written by an agent has to be revised.
- How Kestra looks is decided by the design system: see UI Design System. No color, spacing, radius or component pattern may be invented in feature code.
- A test has to be able to fail for a reason a reviewer would care about: see When a test earns its place. More tests do not make a change safer.
- Build and test only what you touched rather than the whole repo: see What to run after a change.
Comments
The default is no comment. Small methods with accurate names need no explanation, and every comment is a second thing that has to be kept in sync with the code. Over-commenting is easy to slip into: the next line gets narrated, the method name gets restated above the method, and // Step 1 / // Step 2 scaffolding is left behind. None of that will be accepted in review.
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.
- 6d ago First seen · 523 lines · 6,947 tokens per session scan A abc5d003d869
kestra AGENTS.md is an instructions file published in the GitHub repository kestra-io/kestra (27,998 stars, last pushed yesterday), licensed Apache-2.0. It adds 6,947 tokens to every session, about $0.0347 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
workflow AGENTS.md
AGENTS.md instructions for vercel/workflow, covering agent instructions, architecture, core components, workflow execution model and development commands.
decapod CLAUDE.md
Claude Code instructions for DecapodLabs/decapod, covering claude.md - agent entrypoint, orientation & documentation, mandatory prompt safety gate, project context and quick start.
team-shinchan AGENTS.md
Instructions for seokan-jeong/team-shinchan, covering agents.md - team-shinchan agent map, layer architecture, agent registry, call flow and quick reference.
simple-agent-manager AGENTS.md
AGENTS.md instructions for raphaeltm/simple-agent-manager, covering sam — agent supplement, agent configuration cross-reference, skills, review / specialist skills and reference skills.
ultron-control-center CLAUDE.md
Claude Code instructions for SkiTemplar/ultron-control-center, covering claude.md — ultron control center, qué es, build y ejecución — gotchas (causan la mayoría de líos), memoria (sistema propio — no mem0) and routing de skills/agentes.
jarvis-registry copilot-instructions.md
Copilot instructions for ascending-llc/jarvis-registry, a project described as: Connect any AI copilot or autonomous agent to your enterprise tools — through a single, secure MCP/Agent gateway with built-in identity, access control, and full observability.