Prompt Studio v1.0 is live on PyPI · pip install promptstudio-ai

The Lean Prompt
Workbench

We build the tooling and the persona layer that make LLM output shorter, cheaper, and closer to what you actually meant.

Launch StudioInstall from PyPI[X 47.6, Y 57.2]
$pip install promptstudio-ai

Score any prompt, seven ways

7-dimension quality breakdown with letter grade. Clarity, specificity, context, format, mode-fit, token efficiency, constraints.

/A.01

Optimize at the speed of iteration

Rule-based improvement pass adds missing persona, format, examples, constraints. Filler-token compression without semantic loss.

/A.02

75% fewer tokens, measured

Lean persona layer cuts LLM output size, cost, and latency. One SKILL.md, mode-filtered (lite/full/ultra), zero drift across hosts.

/A.03
/D.01 [X 34.6, Y 70.2]

Introducing The
Prompt Studio
Optimization Stack

Analyze scores, token counts, and context-window fit across GPT, Claude, Gemini, Llama, Mistral, and DeepSeek. Optimize with rule-based passes, compress with filler-token stripping. Persist every run to PostgreSQL with pgvector semantic search.

[X 91.7, Y 54.0]
L
/C.02
LEAN
Persona Layer
One SKILL.md that follows your agent everywhere.

Ships as a Claude Code plugin, a Codex adapter, a Copilot CLI plugin, a Devin plugin, a Qoder ruleset, and instruction files for Cursor / Windsurf / Cline / Kiro. Same source of truth, filtered per intensity.

M
/C.03
LEAN-MCP
Prompt-Menu Server
The only prompt-menu MCP for the Lean ruleset in production.

Standalone stdio server for MCP hosts whose only injection point is the prompt menu. Zero drift with the FastAPI adapters · both call the same get_lean_instructions().

[X 71.0, Y 60.4]

Latest

All updates →
Release

Prompt-Studio v1.0: the workbench + the layer, one repo

Analyze, score, optimize, and ship a persona that follows your agent across seven hosts.

Read more →
Lean

Lean intensity levels: lite / full / ultra

One SKILL.md, three payloads. Ultra for long agentic sessions, lite for cost-sensitive calls.

Read more →
Benchmark

Measuring what a persona actually costs

LOC, tokens, USD, latency across five arms. Anthropic + OpenAI + Gemini + Ollama backends.

Read more →
MCP

lean-mcp: same rules, prompt-menu injection

Stdio server for hosts whose only hook is the prompt menu. Zero drift with the FastAPI adapters.

Read more →