Iranti Python Client
Python client for the Iranti REST API.
Installation
pip install iranti
For local development from this repository:
cd clients/python
pip install -e .
Setup
Start the Iranti API server first:
# In your Iranti directory
npm run api
Usage
from iranti import IrantiClient
client = IrantiClient(
base_url="http://localhost:3001",
api_key="replace-with-real-iranti-key"
)
# Or use environment variables
# IRANTI_URL=http://localhost:3001
# IRANTI_API_KEY=replace-with-real-iranti-key
client = IrantiClient()
# Check server is running
print(client.health())
# Write a fact
result = client.write(
entity="researcher/jane_smith",
key="affiliation",
value={"institution": "MIT"},
summary="Affiliated with MIT",
confidence=85,
source="OpenAlex",
agent="my_agent"
)
print(result.action) # created / updated / rejected / escalated
# Ingest raw text
result = client.ingest(
entity="researcher/jane_smith",
content="Dr. Jane Smith has 24 publications and is at MIT CSAIL.",
source="OpenAlex",
confidence=80,
agent="my_agent"
)
print(
f"Extracted: {result.extracted_candidates}, "
f"Written: {result.written}, "
f"Skipped malformed: {result.skipped_malformed}"
)
# Get working memory before a task
brief = client.handshake(
agent="my_agent",
task="Research publication history",
recent_messages=["Starting research..."]
)
print(brief.inferred_task_type)
# Per-turn memory decision (inject only when needed)
turn = client.attend(
agent_id="my_agent",
latest_message="What is my favorite snack?",
current_context="User: What is my favorite snack?\nAssistant:",
entity_hints=["user/main"]
)
print(turn["shouldInject"], turn["reason"])
# Save task progress so a later handshake can recommend resuming it
brief = client.checkpoint(
agent_id="my_agent",
task="Research publication history",
recent_messages=["Still comparing two source timelines."],
checkpoint={
"currentStep": "Review affiliation conflict",
"nextStep": "Write corrected fact",
"openRisks": ["Two sources disagree on start year"],
},
)
if brief.session_recovery and brief.session_recovery.available:
client.resume_session(
agent_id="my_agent",
session_id=brief.session_recovery.session_id
)
session = client.inspect_session("my_agent")
print(session.has_checkpoint)
sessions = client.list_sessions(operator_state="interrupted", sort="operator")
print([(item.agent_id, item.operator_state) for item in sessions])
# Query facts
result = client.query("researcher/jane_smith", "affiliation")
if result.found:
print(result.value)
Error Handling
from iranti import IrantiClient, IrantiAuthError, IrantiValidationError, IrantiError
try:
result = client.write(...)
except IrantiAuthError:
print("Invalid API key")
except IrantiValidationError as e:
print(f"Bad input: {e}")
except IrantiError as e:
print(f"API error: {e}")
Environment Variables
| Variable | Description |
|---|---|
IRANTI_URL |
API server URL (default: http://localhost:3001) |
IRANTI_API_KEY |
API token (keyId.secret) or legacy shared server key |
License
AGPL-3.0-or-later.
1 maintainer
Metadata
1 maintainer
Release files for iranti 0.4.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| iranti-0.4.1.tar.gz | 10.8 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| iranti-0.4.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 21.8 kB
Release files / iranti-0.4.1.tar.gz
| Download URL | iranti-0.4.1.tar.gz |
|---|---|
| Size | 10.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
748b7ec1f08b30d1dcac5688b865c792513496dbb4899f9a2774354c1645a5ba
|
|
BLAKE2b-256 checksum How to use checksums |
97597d8c6ce8629704052e49ac6e7f0efe422be4e62a92a80daa76a7a4c3ee8c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
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 Apr 15, 2026.
Transparency logRelease files / iranti-0.4.1-py3-none-any.whl
| Download URL | iranti-0.4.1-py3-none-any.whl |
|---|---|
| Size | 11.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
ae01f3469dc6f9e974fae2aa702689bc40557f4eabbaead06122dde76296254d
|
|
BLAKE2b-256 checksum How to use checksums |
e005ee2407fe82f9db1e5af4570895a2c0f204029a11c32fdc1e0a4fd0fbb9ee
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
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 Apr 15, 2026.
Transparency log
