aacp-langchain
AACP coordination layer for LangChain multi-agent workflows.
AACP replaces natural language agent-to-agent instructions with typed, validated, deterministic packets. LangChain handles agent lifecycle, memory, and tooling. AACP handles what the agents say to each other.
Install
pip install aacp-langchain
Quick start
from aacp_langchain import AACPOrchestrator
orch = AACPOrchestrator(model="gpt-4.1-mini")
result = orch.run_workflow("payroll", period="2026-03")
print(result.summary())
Measured results
Payroll workflow comparison. gpt-4.1-mini. June 2026.
WITHOUT AACP WITH AACP
Coordination LLM calls: 4 0
Coordination cost: $0.0024 $0.0000
Agent cost: $0.0035 $0.0035
Total cost: $0.0059 $0.0035
Total saving: 18%
Coordination deterministic: NO YES
Schema validated: NO YES
Department day comparison (59 coordination hops)
python3 examples/department_comparison.py --mock
WITHOUT AACP WITH AACP
Coordination LLM calls: 59 0
Coordination cost: $0.0031 $0.0000
Total saving: 18%
Deterministic: NO YES
Without AACP vs With AACP
Without aacp-langchain:
Orchestrator
↓ "Please ask the HR agent to retrieve salary records..."
LLM call ← costs tokens, varies every run
↓ "HR Agent, could you fetch the employee salary data?"
HR Agent
With aacp-langchain:
Orchestrator
↓ FETCH|HR|return:ORCHESTRATOR|p:1|aacp:1.1|res:emp_salary|period:2026-03
HR Agent ✓ validates. $0.00 encoding.
Workflows
# Payroll (5 hops)
result = orch.run_workflow("payroll", period="2026-03")
# IT provisioning / JML onboarding (6 hops)
result = orch.run_workflow(
"it_provisioning",
username="j.smith",
dept="Engineering",
licences=["M365", "Slack"],
)
# Sales qualification (5 hops)
result = orch.run_workflow(
"sales_qualification",
lead_id="L-001",
lead={"budget_gbp": 75000, "timeline_months": 3},
)
Requirements
- Python 3.10+
OPENAI_API_KEYpip install aacp-langchain
Links
- Protocol spec: https://aacp.dev
- Python SDK: https://github.com/MackayAndrew/aacp
- CrewAI integration: https://github.com/MackayAndrew/aacp-crewai
- Community rules (241): https://github.com/MackayAndrew/aacp-community-rules
- IETF Draft: https://datatracker.ietf.org/doc/draft-mackay-aacp/
Licence
MIT
Metadata
Release files for aacp-langchain 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| aacp_langchain-0.1.0.tar.gz | 19.6 kB | Without provenance | Details for aacp_langchain-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| aacp_langchain-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for aacp_langchain-0.1.0-py3-none-any.whl |
Total release size: 42.1 kB
Release files / aacp_langchain-0.1.0.tar.gz
| Download URL | aacp_langchain-0.1.0.tar.gz |
|---|---|
| Size | 19.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
9c66d2816a876fe7aa52a4832883777a9a78a60bf1e48f315f455bd8a96dc13d
|
|
BLAKE2b-256 checksum How to use checksums |
86a277ef3406f2a4346a83b8295c305ad14af5b984498a7c9a9aa694fc9932b4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.5
|
Release files / aacp_langchain-0.1.0-py3-none-any.whl
| Download URL | aacp_langchain-0.1.0-py3-none-any.whl |
|---|---|
| Size | 22.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
aed5ccb7d8b5cf770bb4da594cbb92a065aff89609791808656e5d99caf0d930
|
|
BLAKE2b-256 checksum How to use checksums |
512601bf484c747696cab62084f3cb5837291ac338d67857590f5f5aba5b04a0
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.5
|
Security for aacp-langchain 0.1.0
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.
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.
To learn more about protecting yourself from malicious actors, visit our package security documentation.