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 skills/zenml-io/kitaru/kitaru-importer-developmentnpx skills add zenml-io/kitaru --skill kitaru-importer-developmentgit clone --depth 1 https://github.com/zenml-io/kitaruWhat 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 | $0.00048 | $0.01074 |
| Opus 5 | $0.00024 | $0.00537 |
| Sonnet 5 | $0.00010 | $0.00215 |
| Haiku 4.5 | $0.00005 | $0.00107 |
Grade A, and why
kitaru-importer-development 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 yesterday.
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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Kitaru Importer Development
Read AGENTS.md, plugins/AGENTS.md, plugins/DEVELOPMENT.md, and src/kitaru/task/importer.py. Treat the task module as the executable importer contract. Read the JSONL importer first for the smallest example, then the closest provider importer and its fixtures. Load the same-name kitaru-dev repo skill for the current host for the general command and PR workflow.
First decide whether the request needs a reusable package. A one-off conversion to Kitaru JSONL or a self-contained registered script may be smaller. Registration stores source metadata; it does not upload or vendor a package.
Implement the parser contract
Implement parse(payload: bytes, params: dict[str, Any]) as an iterator of ImportedSession or ImportFailure.
- Choose and document the source-to-session boundary.
- Derive a stable source
external_id; Kitaru deduplicates using the importer provider and this ID. - Preserve source inputs and outputs. Populate selectors, models, tokens, costs, reasoning, framework, attributes, and metadata only when the export supports them without guessing.
- Yield
ImportFailurefor an isolated bad record. An exception raised while starting or advancing the parser ends the import task. - Preserve valid node ordering and graph relationships. Use either nested nodes or the flat indexed form accepted by
flatten_nodes; follow that function's validation rules. - Validate provider parameters and keep normalization or grouping provenance in metadata when it changes how the source is interpreted.
Use representative provider exports as fixtures. Cover malformed records, missing or duplicate IDs, grouping, ordering, parent links, model and tool normalization, reasoning visibility, selector escaping, and stable re-import behavior as applicable.
Package, register, and version
An importer distribution lives under plugins/packages/<slug>-importer/, with its source, pyproject.toml, changelog, and focused tests under plugins/tests/importers/. Export parse through the package __all__. Add the package to plugins/README.md, release/release-units.toml, the exact inventory in tests/scripts/test_release_units.py, and plugins/uv.lock. A non-default package also declares tool.kitaru.artifact.import-module so artifact smoke can import it without a default-catalog entry.
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.
- yesterday First seen · 59 lines · 48 tokens per session scan A 5d375b7e9ad6
kitaru-importer-development is a skill published in the GitHub repository zenml-io/kitaru (270 stars, last pushed 4d ago), licensed Apache-2.0. It adds 48 tokens to every session and 1,074 once invoked, about $0.0002 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 skills, from other repositories
migrating-world-v4-to-v5
Upgrades a custom Workflow SDK World implementation from the v4 spec to v5. Use when a package implements the World interface from @workflow/world and is moving to 5.x — event IDs that are ULIDs rather than slot positions, Event id is not slot-numbered at replay time, a specVersion the runtime refuses, writeToStream /…
field-descriptions
Write or review Pydantic Field(description=...) text for ZenML stack component configs. Use when adding, editing, or reviewing config fields in src/zenml/integrations//flavors/ or any StackComponentConfig subclass, or when scripts/validatedescriptions.py fails.
workflow-init
Install and configure Vercel Workflow SDK before it exists in nodemodules. Use when the user asks to "install workflow", "set up workflow", "add durable workflows", "configure workflow sdk", or "init workflow" for Next.js, Express, Hono, Fastify, NestJS, Nitro, Nuxt, Astro, SvelteKit, or Vite.
experiment-tracking-swanlab
Provides guidance for experiment tracking with SwanLab. Use when you need open-source run tracking, local or self-hosted dashboards, and lightweight media logging for ML workflows.
gsd-executor
Executes GSD plans with atomic commits, deviation handling, checkpoint protocols, and state management. Spawned by execute-phase orchestrator or execute-plan command.
mlflow
Track ML experiments, manage model registry with versioning, deploy models to production, and reproduce experiments with MLflow - framework-agnostic ML lifecycle platform.