mirror of
https://github.com/magnus919/agent-skills.git
synced 2026-09-12 12:06:29 +03:00
Magnus Hedemark
95046675bc
Greenfield SkillOpt: 3 epochs for a Stanford DSPy compiler skill. DSPy is a fundamentally different paradigm from chain/RAG frameworks. Epoch 1 — Prominence: - Hard-gate blockquote: 'DSPy is NOT a chain framework' - Core Paradigm section with runnable code example early Epoch 2 — Decision Guidance: - Framework Routing Guide (DSPy vs LlamaIndex vs LangChain vs LangGraph) - Where to Start table mapping entry points - Troubleshooting table with reference links Epoch 3 — Pattern Expansion: - Optimizer selection cheat sheet from official docs - Caching, compilation cost management, save/load - FAQ covering paradigm confusion, errors, deployment 12 files: SKILL.md, 7 references, 3 templates, 1 script. v1.0.0 -> v1.0.3 across 3 epochs. All API surfaces validated against dspy.ai official docs — optimizer selection guide, caching, core modules, FAQ. Signed-off-by: Jasper <jasper@montcastle.bitches>
26 lines
937 B
Python
26 lines
937 B
Python
#!/usr/bin/env python3
|
|
"""Multi-step reasoning with DSPy — decompose, answer, synthesize."""
|
|
|
|
import dspy
|
|
|
|
lm = dspy.LM("openai/gpt-4o-mini")
|
|
dspy.configure(lm=lm)
|
|
|
|
class MultiStepQA(dspy.Module):
|
|
def __init__(self):
|
|
self.decompose = dspy.ChainOfThought("question -> sub_questions")
|
|
self.answer = dspy.ChainOfThought("sub_question -> answer")
|
|
self.synthesize = dspy.ChainOfThought("answers -> final_answer")
|
|
|
|
def forward(self, question):
|
|
sub_qs = self.decompose(question=question).sub_questions
|
|
answers = []
|
|
for q in sub_qs:
|
|
answers.append(self.answer(sub_question=q).answer)
|
|
combined = "\n".join(f"Q: {q}\nA: {a}" for q, a in zip(sub_qs, answers))
|
|
return self.synthesize(answers=combined)
|
|
|
|
program = MultiStepQA()
|
|
result = program(question="What are the benefits of using DSPy for prompt optimization?")
|
|
print(f"Final Answer: {result.final_answer}")
|