djangorestframework-pydantic-ai
Expose djangorestframework-services
services and selectors as a Pydantic-AI toolset, so a
plain pydantic_ai.Agent can call them as tools — no MCP server and no AG-UI
bridge in the path.
Every tool call routes through drf-services' transport-neutral surface
(dispatch_spec plus its off-HTTP helpers), so a tool call runs the same input
validation, the same permission_classes — both class-level has_permission
and object-level has_object_permission on the resolved row — and the same
serializer rendering your DRF views apply, just without the HTTP hop.
Install
pip install djangorestframework-pydantic-ai
It depends only on djangorestframework-services and pydantic-ai-slim. A model
provider is pulled in separately, the usual Pydantic-AI way (e.g.
pip install "pydantic-ai-slim[anthropic]").
Quickstart
from pydantic_ai import Agent
from rest_framework_pydantic_ai import AgentDeps, SpecToolset
toolset = SpecToolset(
{
"list_orders": orders_selector_spec, # SelectorSpec -> read-only tool
"create_order": create_order_spec, # ServiceSpec -> mutation tool
}
)
agent = Agent("anthropic:claude-opus-4-8", deps_type=AgentDeps, toolsets=[toolset])
result = await agent.run(
"create an order of 3 widgets",
deps=AgentDeps(user=request.user),
)
The agent acts as deps.user: each call builds an off-HTTP request/view context,
enforces the spec's permission_classes (class- and object-level),
dispatches the spec, and renders the result through the spec's serializer. List
selectors gain page / limit / order tool args, and return one page —
{"items": [...], "page": 1, "totalPages": N, "hasNext": bool} — so a model
reading a collection can tell whether it saw all of it.
Failures map onto the model loop so the agent self-corrects: invalid input, a bad
page / limit / order value, and unexpected (hallucinated) arguments come
back as ModelRetry; a business error or missing row becomes a readable
ToolFailed, which the model sees as a failed call rather than a successful one
carrying prose, and a call an Affordance refuses names the rule as well as the
reason (The books are closed. (code: books_closed)); a denied permission raises PermissionDenied, which aborts
the run under a plain pydantic_ai.Agent. A host that installs a tool-failure
policy — django-pydantic-agent's build_agent does by default — turns that
denial into a failed-tool result instead, so the run continues; the call is
still denied either way. Unexpected arguments are rejected by default — pass
unknown_arguments= to SpecToolset (IGNORE / PASSTHROUGH) to change that.
A tool whose Affordance condition is unmet right now, such as posting to a closed
period, is left out of that step's tool list, and the step's instructions name it
with its reason so the model can tell the user why. Permissions do not filter the
list: a denied tool is still offered, and the denial comes on the call.
Every line of the instructions block is a field of AgentConventions, the
heading of the unavailable list included, and so are a handle field's fallback
description, the sentence scoping a paged tool's read-shaping parameters, and the
missing-argument retry. SpecToolset(specs, conventions=AgentConventions(pagination=...))
rewords one line and leaves the rest as they are. The toolset still decides when
each line is said, and None drops a line. Argument descriptions, such as those
of limit and page, and the toolset's other retries stay fixed.
See the documentation for the full reference.
License
MIT
1 maintainer
Metadata
1 maintainer
Release files for djangorestframework-pydantic-ai 0.35.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| djangorestframework_pydantic_ai-0.35.0.tar.gz | 335.4 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| djangorestframework_pydantic_ai-0.35.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 410.5 kB
Release files / djangorestframework_pydantic_ai-0.35.0.tar.gz
| Download URL | djangorestframework_pydantic_ai-0.35.0.tar.gz |
|---|---|
| Size | 335.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f2fa819f3554226f4b3b6e08dad61b503f7a39282b79c3077c48c27a1f7e33b2
|
|
BLAKE2b-256 checksum How to use checksums |
95d26b621ff941336baea320306f426b5a4670bd1591f866151989423f59dab8
|
| 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 7, 2026.
Transparency logRelease files / djangorestframework_pydantic_ai-0.35.0-py3-none-any.whl
| Download URL | djangorestframework_pydantic_ai-0.35.0-py3-none-any.whl |
|---|---|
| Size | 75.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
6cb82d4effa15cc854482730b5e7ca142a3ea8698b1b45a4ade582fd5a1bba26
|
|
BLAKE2b-256 checksum How to use checksums |
ab2d0c95c25845a0aa538ca95cc3f11b7b93f2a4490cfa9f29b5c6618fe3e924
|
| 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 7, 2026.
Transparency log
