Skip to main content

vekna

Run coding agents as rituals: ordinary Python programs whose steps you control and whose agent calls happen inside those steps. Agents run permissively within a step; determinism lives at the step boundaries.

Documentation is at vekna.fancysnake.dev.

Install

pip install vekna

Python 3.11+. Testing your rituals needs the trial extra: pip install 'vekna[trial]' — quoted, because zsh globs the brackets.

A ritual

Put a rituals.py in your project — or a rituals/ package, split however you like, once one file stops being enough:

from typing import Annotated

from pydantic import BaseModel, Field

from vekna.folio.coding import coding
from vekna.folio.shell import shell
from vekna.lexicon import Done, ritual, step


class FixTests(BaseModel):
    # A retry budget counts down to zero, so the CLI rejects a negative one
    # rather than letting `--bound -1` run until the step backstop.
    bound: Annotated[int, Field(ge=0)] = 3


class Attempt(BaseModel):
    left: int


class Verdict(BaseModel):
    outcome: str


# The return annotation is the step's exits: back to `fix`, or finished with a
# `Verdict`. mypy checks every return against it, so a mis-wire fails the type
# check rather than a cast.
@step
async def fix(state: Attempt) -> Attempt | Done[Verdict]:
    result = await shell("mise run test:py")
    if result.exit_code == 0:
        return Done(Verdict(outcome="green"))
    if state.left <= 0:
        return Done(Verdict(outcome="gave up"))
    await coding(f"The test suite fails:\n{result.stdout}\nFix it.")
    return Attempt(left=state.left - 1)


# `def`, not `async def`: naming the first step has nothing to await. A step or
# entrypoint is written whichever way its body needs.
@ritual("fix_tests")
def fix_tests(components: FixTests) -> Attempt:
    return Attempt(left=components.bound)

Then cast it:

vekna cast fix_tests --bound 5

Output streams live as a tree of rites — one node per step, one nested under it per medium call, with the agent's own output indented beneath. The last line is the cast's result, as JSON:

result: {"outcome":"green"}

Claude Code plugin

The ritual-scribe skill teaches Claude Code to write rituals. Install it from this repo's marketplace:

claude plugin marketplace add fancysnake/vekna
claude plugin install ritual-scribe@vekna

Commands

Command What it does
vekna cast <ritual> [--<component> value …] Run a ritual from rituals.py
vekna cast --prompt "<text>" One-shot cast on the coding medium, no rituals.py needed
vekna rituals list Every ritual and the options it takes
vekna rituals show <ritual> A ritual's components and its step graph

Architecture

GLIMPSE layering, enforced by import-linter. See docs/architecture.md and docs/README.md for where the ideas live.

Development

mise run test:py     # all tests
mise run check:py    # the loop while you work: format, lint, tests
mise run fullcheck   # the gate before you push: adds diff-coverage and tingle

Licence

BSD-3-Clause.

Metadata

Release files for vekna 0.15.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for vekna 0.15.0
File Size Uploaded Provenance Actions
vekna-0.15.0.tar.gz 125.7 kB With provenance Details for vekna-0.15.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for vekna 0.15.0
File Interpreter ABI Platform Actions
vekna-0.15.0-py3-none-any.whl Python 3 none any Details for vekna-0.15.0-py3-none-any.whl

Total release size: 282.6 kB

Release files / vekna-0.15.0.tar.gz

Download URL vekna-0.15.0.tar.gz
Size 125.7 kB
Tags Source
SHA-256 checksum
How to use checksums
e18c19903c44a791cc439c59becc022cb48051e0cd0f6bd234ce764ac0cee344
BLAKE2b-256 checksum
How to use checksums
f5adefbbdc5129c39e4cc9ed4015bd2d9b5b1ecd0b3c34384cdfd3d860bf3435
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 26, 2026.

Transparency log

Release files / vekna-0.15.0-py3-none-any.whl

Download URL vekna-0.15.0-py3-none-any.whl
Size 156.9 kB
Tags Python 3
SHA-256 checksum
How to use checksums
f5e52b345cd91a39ebf5dcf14692011f6a68ec42c7822f4999ced37d8c47f705
BLAKE2b-256 checksum
How to use checksums
78d2ac62df6d36755b0ece93e5281b18c3bcf8154a8dc9153ed9deeccfd65ef9
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 26, 2026.

Transparency log

Security for vekna 0.15.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for vekna-0.15.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 26, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page