Skip to main content

zizkadb-crewai

PyPI: zizkadb-crewai 0.1.3 · requires zizkadb-sdk>=0.3.0

Self-hosted, open-source observability and causal debugging for CrewAI.

A CrewAI run is normally a black box: when the output is wrong, it's hard to see which task or agent caused it. zizkadb-crewai logs every step of a crew run to ZizkaDB with a parent_id link, so a run becomes a causal tree — and one call, db.why(), walks that tree back to the root cause.

It doesn't only show what happened; it explains why it happened.

CrewAI run ──▶ ZizkaDBCrewLogger ──▶ ZizkaDB (causal store) ──▶ db.why() / dashboard

Install

Both packages are published on PyPI — no git URLs needed:

pip install zizkadb-sdk zizkadb-crewai crewai

Usage

ZizkaDBCrewLogger logs the crew lifecycle. Events auto-chain through the logger's last_event_id, so you can omit parent_id to link each event to the previous one.

Method Event type Purpose
await log_kickoff(goal, **extra) crew_kickoff Start of the run (root of the chain)
await log_task(description, *, parent_id=None, **extra) crew_task One node per task
await log_output(output, *, parent_id=None, **extra) crew_output Final result
from zizkadb import ZizkaDB
from zizkadb_crewai import ZizkaDBCrewLogger

# OSS-first: default to a local self-hosted instance; pass api_key only for remote/cloud.
async with ZizkaDB(host="http://localhost:8000") as db:
    logger = ZizkaDBCrewLogger(db=db, agent="research-crew")

    await logger.log_kickoff(goal="Research causal logging for AI agents")
    result = await crew.kickoff_async()
    for task_output in result.tasks_output:                    # one causal node per task
        await logger.log_task(
            description=task_output.description,
            output=str(task_output.raw)[:2000],
        )
    await logger.log_output(str(result))

    # The payoff — explain WHY the crew produced its output:
    (await db.why(logger.last_event_id)).print()

db.why() reconstructs the chain:

crew_kickoff → crew_task (Researcher) → crew_task (Writer) → crew_output

Runnable example

A complete two-agent crew (Researcher → Writer) with the why() proof lives at examples/crewai-agent/. It targets a self-hosted ZizkaDB (http://localhost:8000) — no cloud signup required.

Monorepo dev install

pip install -e sdk/python -e integrations/crewai

License

zizkadb-crewai and zizkadb-sdk are licensed under AGPL-3.0.

Metadata

Release files for zizkadb-crewai 0.1.3

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

Source distribution (sdist)

Source distribution for zizkadb-crewai 0.1.3
File Size Uploaded Provenance
zizkadb_crewai-0.1.3.tar.gz 3.8 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for zizkadb-crewai 0.1.3
File Interpreter ABI Platform
zizkadb_crewai-0.1.3-py3-none-any.whl Python 3 none any Details

Total release size: 7.9 kB

Release files / zizkadb_crewai-0.1.3.tar.gz

Download URL zizkadb_crewai-0.1.3.tar.gz
Size 3.8 kB
Tags Source
SHA-256 checksum
How to use checksums
83b78992622571be6d2519eb22a7a11800d95b949dc915c88ee9cbd3d1b2dd62
BLAKE2b-256 checksum
How to use checksums
3f256a8b3d1a349474091345fbae76c73b0f045bc96d65095a50b5280111268c
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.13.9

Release files / zizkadb_crewai-0.1.3-py3-none-any.whl

Download URL zizkadb_crewai-0.1.3-py3-none-any.whl
Size 4.1 kB
Tags Python 3
SHA-256 checksum
How to use checksums
9b96641691fba4ade0f39439f3b24f39d2775b48d7dc2f5ca3c94097f68b9c20
BLAKE2b-256 checksum
How to use checksums
6d8b8b7628694e39c493b25fa918e89b386ccf6f4a918184b53efcafbedeec8b
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.13.9

Security for zizkadb-crewai 0.1.3

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.

No provenance found

None of the files in this release were uploaded with provenance metadata.

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.

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