skills.rest
agenticgogol
agenticgogolCOMMUNITY
Explore all skills by this author →
GitHub

agent-dspy

Implements DSPy programs with Signatures, Modules, and optimizers against a verified LLM provider.

1|9|Updated Jul 11, 2026
Search Tags:#python#prompt optimization#llm#agent#signature#dspy#programmatic prompting
One-click install
npx skills add https://github.com/agenticgogol/Edureka_Coding_Agent_Enabled_Demo_11Jul --skill agent-dspy-agenticgogol
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-dspy
Source: https://github.com/agenticgogol/Edureka_Coding_Agent_Enabled_Demo_11Jul/tree/main/.claude/skills/agent-dspy
Command: npx skills add https://github.com/agenticgogol/Edureka_Coding_Agent_Enabled_Demo_11Jul --skill agent-dspy-agenticgogol

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires dspy, and includes references (resource) components.

What problem does it solve? Building agents with DSPy is risky because its public API (Signatures, Modules, Optimizers/Teleprompters) changes frequently and breaks backward compatibility across versions. This Skill enforces a research-first, spike-first workflow so DSPy code is validated against the currently installed version before being wired into a project. ## Core Features & Use Cases - Research-first workflow: Requires fetching DSPy's current official docs before writing any code, since the bundled reference is a starting sketch, not ground truth. - Reference implementation: Adapts a minimal Signature + Module + optimizer (BootstrapFewShot-style) example from references/basic_dspy_program.py. - Real-provider enforcement: Routes LM calls through DSPy's LM configuration backed by a verified API key — no mock mode, since require-api-key has already validated a working key. - Use Case: When a project brief explicitly names DSPy for programmatic prompting or prompt optimization, use this Skill to build a run_program(input) -> output entrypoint that is spike-tested standalone before integration. ## Quick Start Use the agent-dspy skill to build a DSPy question-answering program with a Signature, ChainOfThought module, and a verified Anthropic API key.

Frequently Asked Questions about agent-dspy

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a DSPy program with Signatures and Modules?▼

Define a dspy.Signature subclass with InputField and OutputField declarations, then wrap it in a dspy.Module using a predictor like dspy.ChainOfThought. Expose a single entrypoint such as run_program(question) that returns the module's output.

How do I configure DSPy with an Anthropic API key?▼

Create an LM instance with dspy.LM using a provider-prefixed model string like anthropic/claude-sonnet-5 and pass your API key, then call dspy.configure(lm=lm). The exact configuration API varies by DSPy version, so verify against the installed version's docs first.

Why does DSPy code break after upgrading the library?▼

DSPy's public API has broken backward compatibility across versions, including changes from dspy.OpenAI to dspy.LM and renamed teleprompters. Always check the currently installed version's official docs and spike-test your code standalone before integrating it.

Can I use DSPy without an API key or in mock mode?▼

No. This workflow has no mock mode — the configure_lm function raises a RuntimeError if the API key is empty. A require-api-key gate verifies a working provider key with an actual call before any DSPy code runs.

What is the difference between DSPy optimizers and teleprompters?▼

Optimizers (formerly called teleprompters) like BootstrapFewShot automatically generate and select few-shot examples to improve a program's prompts. Naming and APIs for these have changed across DSPy versions, so confirm the current class names in the official docs.

Related Skills

Explore More Software Engineering Skills→
ArdurAIArdurAI
1

god-ai-prompting

Design, test, and optimize prompts for large language models with DSPy.

Official
Advanced
soniqosoniqo
1.1k

test

Run Swift unit or end-to-end test suites against a local codebase.

Official
Basic
Araara7Araara7

dspy

Build declarative LM programs with automatic prompt optimization and modular pipelines.

Community
Advanced
eliottbusinesseliottbusiness

dspy

Compile declarative language model calls into self-improving pipelines.

Community
Advanced