agentsope/SkillAlchemy / skills/agentsop-dspy
agentsop-dspy
VerifiedOperating SOP for DSPy (Stanford NLP) — the declarative framework for "programming, not prompting" language models. Activate when the user says any of: "use DSPy", "compile a prompt", "optimize prompts/programs", "MIPRO/MIPROv2", "BootstrapFewShot", "GEPA", "Signatures + Modules", "teleprompter", "auto-tune prompts for a different LM", or whenever a brittle hand-crafted prompt pipeline needs to be turned into a *compiled*, measurable, swappable program. Do NOT activate for one-shot prompt tweaks, no-metric exploratory work, or pipelines where prompts must remain human-authored verbatim — use raw prompting or LangChain templates instead.
| File | Kind | Size |
|---|---|---|
| README.md | markdown | 2.4 KB |
| SKILL.md | skill md | 28.8 KB |
| intermediate/operation_candidates.json | config | 12.3 KB |
| references/R1-architecture.md | markdown | 5.9 KB |
| references/R2-sop-workflow.md | markdown | 5.5 KB |
| references/R3-dilemma-cases.md | markdown | 7.5 KB |
| references/R4-anti-patterns.md | markdown | 4.9 KB |
| references/R5-ecosystem-context.md | markdown | 7.3 KB |
README.md
View on GitHub# DSPy SOP Skill — Package README
This package distills DSPy (Stanford NLP) into an actionable SOP skill for coder-agents. It is **not** API documentation — it is a *decision system* for *when* and *how* to apply DSPy primitives in real engineering work.
## Contents
```
dspy-sop-skill/
├── SKILL.md # main skill — 7 sections, frontmatter, 4 dilemma cases
├── README.md # this file
├── references/
│ ├── R1-architecture.md # Signatures / Modules / Teleprompters / Compile mental model
│ ├── R2-sop-workflow.md # 3-stage gate workflow with exit criteria
│ ├── R3-dilemma-cases.md # 4 extended dilemma cases with citations
│ ├── R4-anti-patterns.md # 10 anti-patterns + boundaries
│ └── R5-ecosystem-context.md # DSPy vs LangChain/LangGraph/LlamaIndex/Guidance
└── intermediate/
└── operation_candidates.json # structured Trigger/Action/Output/Evidence operations
```
## How to use
A coder-agent should load `SKILL.md` when any activation trigger in §1 fires. The `references/` files are deeper drill-downs for sub-tasks (e.g. when the user asks "explain GEPA vs MIPROv2", the agent loads `R3-dilemma-cases.md` Case C and §4.1 of SKILL.md).
The `intermediate/operation_candidates.json` is machine-readable for the meta-pipeline that synthesizes across the 7-project landscape (LangGraph, LlamaIndex, DSPy, CrewAI, vLLM, Aider, Dify).
## Provenance & quality notes
- **No fabrication.** Every claim has an inline citation: `[dspy.ai/...]`, `[arxiv.org/...]`, `[github.com/stanfordnlp/dspy/issues/...]`, or named third-party (Databricks blog, Miyamura substack, langwatch comparison).
- **Real dilemma cases.** Cases A–D draw from community-reported issues (#1596, #1970, #1078) and documented production case studies (JetBlue, Databricks, Haize Labs).
- **Coverage caveats**:
- Discord war stories are paraphrased from community write-ups; raw Discord transcripts were not fetched (private).
- The DSPy v3.x API surface (post v3.2.1) was sampled via docs as of research date 2026-05-19; minor API drift possible.
- GEPA is recent (ICLR 2026 oral) — practitioner war stories are thinner than for MIPROv2.
## Version
- **Skill version**: 0.1.0
- **DSPy version referenced**: v3.2.1 (per GitHub README at research time)
- **Last research**: 2026-05-19
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |