beunec-asps
Agentic-System-Prompt-as-a-Skill™ (ASPS™) — A lightweight, zero-dependency Python framework for deterministic skill construction in agentic systems.
An LLM does not "have skills." It has parameters. ASPS™ provides the infrastructure that makes an agentic system skillful.
Install
pip install beunec-asps
Quick Start
from beunec_asps import (
ASPSBuilder,
create_heuristic,
create_pseudonym_protocol,
NetworkTopologies,
HEURISTIC_LIBRARIES,
)
# Build a skill in 4 lines
skill = (
ASPSBuilder(name="Stock Analyst", domain="Equity Research", description="...")
.distill(HEURISTIC_LIBRARIES["financial_stock_analyst"])
.reinforce(
pseudonym_protocol=create_pseudonym_protocol(
identity="CFA Charterholder",
persona="A disciplined equity research analyst.",
),
guardrail_presets=["standard", "financial"],
)
.network(NetworkTopologies.hub_and_spoke(
orchestrator_label="Lead Analyst",
spokes=[{"label": "Market Data API", "node_type": "api", "capabilities": ["quotes"]}],
))
.compile()
)
# Use the compiled system prompt with any LLM
print(skill.compiled_system_prompt)
Pre-Built Templates
from beunec_asps.templates import ASPS_TEMPLATES
# 7 ready-to-use skill templates
skill = ASPS_TEMPLATES["financial_stock_analyst"]()
skill = ASPS_TEMPLATES["full_stack_developer"]()
skill = ASPS_TEMPLATES["scientific_researcher"]()
skill = ASPS_TEMPLATES["content_creator"]()
skill = ASPS_TEMPLATES["private_equity_analyst"]()
skill = ASPS_TEMPLATES["financial_investment_analyst"]()
skill = ASPS_TEMPLATES["academia_professor"]()
The Three Techniques
| Layer | Technique | Purpose |
|---|---|---|
| 1 | ASD™ (Agentic Skill Distillation) | Extract expert heuristics → deterministic instruction chains |
| 2 | ASR™ (Agentic Skill Reinforcement) | Behavioral checkpoints, pseudonym protocols, ICRL, guardrails |
| 3 | ANS™ (Agentic Network System) | Wire skills into governed multi-agent network topologies |
Works With Everything
beunec-asps is zero-dependency and produces plain strings. It works alongside — never conflicts with:
- LangChain / LangGraph — use the compiled prompt as your agent's system message
- OpenAI SDK — pass
skill.compiled_system_promptas the system message - Anthropic SDK — same
- AutoGen / CrewAI — use as the internal prompt for any agent in your graph
- Any LLM — it's just a string
Custom Skills
from beunec_asps import ASPSBuilder, create_heuristic, create_pseudonym_protocol, NetworkTopologies
skill = (
ASPSBuilder(name="My Expert", domain="My Domain", description="What it does")
.distill([
create_heuristic(name="Step 1", instruction="Do this first."),
create_heuristic(name="Step 2", instruction="Then do this."),
])
.reinforce(
pseudonym_protocol=create_pseudonym_protocol(
identity="Domain Expert",
persona="An experienced professional.",
),
guardrail_presets=["standard"],
)
.network(NetworkTopologies.pipeline(stages=[
{"label": "Agent A", "node_type": "agent", "capabilities": ["analyze"]},
{"label": "Agent B", "node_type": "agent", "capabilities": ["synthesize"]},
]))
.compile()
)
License
MIT — © 2025 Beunec Technologies, Inc.
ASPS™, ASD™, ASR™, ANS™ are trademarks of Beunec Technologies, Inc.
Release files for beunec-asps 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 | |
|---|---|---|---|
| beunec_asps-0.1.0.tar.gz | 27.0 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| beunec_asps-0.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 53.8 kB
Release files / beunec_asps-0.1.0.tar.gz
| Download URL | beunec_asps-0.1.0.tar.gz |
|---|---|
| Size | 27.0 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
ff75a63b0b54b52f0fe6de2f5dd311d1d6c080ecfc9285b10e43d1483ff3b413
|
|
BLAKE2b-256 checksum How to use checksums |
d42d49d6ccc824b8f09a4231d4236a046dc18bfb11ab7473b59eec959bf03447
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.5
|
Release files / beunec_asps-0.1.0-py3-none-any.whl
| Download URL | beunec_asps-0.1.0-py3-none-any.whl |
|---|---|
| Size | 26.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
c5a9239e9dcdf5c90bbfaad120cc2724962367713326a9cf5c828ac8d8e76c74
|
|
BLAKE2b-256 checksum How to use checksums |
b91dab27cdff6e5dc36ccf34d49c0882816c99987c0a8d4926232dcc5084b10a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.5
|