A previous release had provenance metadata, but this release does not. View details.
crewai-openterms
CrewAI adapter for OpenTerms — wraps any callable so each invocation emits a signed ORS v0.1 receipt before the tool runs.
Depends on openterms-py>=0.1.0 for canonicalization, signing, and the HTTP client. One signing path is shared with langchain-openterms.
Install
pip install crewai-openterms
CrewAI itself is not a hard runtime dependency. CrewAI's BaseTool API has shifted repeatedly across the 0.x line (constructor signatures, the _run-vs-run hook, pydantic model versions), so this adapter wraps a plain callable — the function-tool pattern, which is stable across CrewAI versions. Plug the wrapped function into CrewAI however your project already does (Tool(name=..., func=...), the @tool decorator, or as the body of a BaseTool._run).
Quickstart
from openterms import IngestClient, generate_keypair
from openterms_crewai import OpenTermsToolConfig, wrap_tool
sk, _ = generate_keypair()
client = IngestClient(
base_url="http://localhost:3000",
workspace_id="00000000-0000-4000-8000-0000000000aa",
key_id="my-key",
private_key=sk.private_bytes_raw(),
agent_id="crew-agent",
)
config = OpenTermsToolConfig(
client=client,
agent_id="crew-agent",
terms_url="https://example.com/terms",
terms_hash="a" * 64,
emit_post_action=True,
)
def search(query: str) -> str:
"""Pretend to look something up."""
return f"results for: {query}"
wrapped_search = wrap_tool(search, config=config, tool_name="search")
# Plug into CrewAI:
# from crewai.tools import Tool
# tool = Tool(name="search", func=wrapped_search, description="...")
Decorator form:
from openterms_crewai import openterms_tool
@openterms_tool(config, tool_name="search")
def search(query: str) -> str:
return f"results for: {query}"
See examples/quickstart.py for a runnable copy.
Configuration
OpenTermsToolConfig carries:
client— theopenterms.IngestClient.agent_id— placed on every receipt.terms_url,terms_hash— terms binding for the receipts.emit_post_action— whenTrue, emit a second receipt after the tool returns withpost_state_hash= SHA-256 of the stringified result. DefaultFalse.strict— whenTrue, re-raiseopenterms.IngestError; defaultFalse(failures logged, tool still runs).
What this adapter does
- Wraps a callable. Each invocation emits a signed ORS receipt with
action_type="tool_call"andaction_context = {tool_id, args}. Argument introspection is best-effort (inspect.signaturefirst, positional fallback for builtins / C callables). - Optionally emits a post-action receipt sharing the same
receipt_idas the pre-action receipt. - Logs and continues on ingest failures by default. The tool's return value is always passed through.
What this adapter does NOT do
- It does not depend on CrewAI at runtime. Tests run without
crewaiinstalled. The wrapper is callable-based. - It does not run an ingest service. You must point
IngestClientat a running OpenTerms API. - It does not handle auth, host a JWKS, or manage keys. See
opentermsfor the underlying scope. - It does not buffer or retry. A failed emission is logged once (or re-raised if
strict=True). - It does not subclass CrewAI's
BaseTool. That subclass tends to break across CrewAI minor versions. If you need aBaseToolsubclass, build one in your project that calls into awrap_tool-produced callable from inside_run.
Repository
Source and issue tracker: jstibal/openterms.
License
Apache-2.0.
Metadata
Release files for crewai-openterms 1.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Loss of provenance
These files do not have provenance metadata, despite files in a prior release including this data.
We recommend further investigation before installing these files. View details.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| crewai_openterms-1.0.1.tar.gz | 6.8 kB | Without provenance | Details for crewai_openterms-1.0.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| crewai_openterms-1.0.1-py3-none-any.whl | Python 3 | none | any | Details for crewai_openterms-1.0.1-py3-none-any.whl |
Total release size: 12.6 kB
Release files / crewai_openterms-1.0.1.tar.gz
Loss of provenance
This file belongs to a release without provenance metadata, despite a prior release including this data.
We recommend further investigation before installing this file. View details.
| Download URL | crewai_openterms-1.0.1.tar.gz |
|---|---|
| Size | 6.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
ee83045c0874796cb5b52a8a08cd2f124a53f2fc6bb37491b86fa48c0c4ca9b9
|
|
BLAKE2b-256 checksum How to use checksums |
deb98f27a83796e8820d77d0c2f5c0124225366e8dd0bc234ef0b4387d962958
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.3
|
Release files / crewai_openterms-1.0.1-py3-none-any.whl
Loss of provenance
This file belongs to a release without provenance metadata, despite a prior release including this data.
We recommend further investigation before installing this file. View details.
| Download URL | crewai_openterms-1.0.1-py3-none-any.whl |
|---|---|
| Size | 5.8 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
7ce7dd462f534d6b432c2373a00ec0f992e58d299af8cbd0d8286c2ea59c870b
|
|
BLAKE2b-256 checksum How to use checksums |
68027e68b23fe1b0f236048c765fad81117954e8f87574f217235c8883f60a1b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.3
|
Security for crewai-openterms 1.0.1
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.
Loss of provenance
A previous release of these files had provenance metadata, but this release does not.
While this could reflect a routine change by the project maintainers, it may also indicate that the release files have been compromised.
Release comparison
- Removed 0.4.0 2 files With provenance
- Added 1.0.1 2 files Without provenance
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.