Optimizes and rewrites prompts using proven prompt engineering techniques: draft diagnosis, canonical structure, technique selection by task type and target model (Claude 4.x, OpenAI, Gemini), with worked examples, a quality checklist, and an evaluation guide.
★not rated 10▲
+1 4mo agoA
tokens not measured
originalMIT
Prompt engineering workbench with a Lean persona layer that cuts LLM output size, cost, and latency. Injects the Lean ruleset every session via SessionStart + SubagentStart hooks.
Clarify and optimize prompts for GPT-6 Astra, GPT-5.6, Claude Sonnet/Opus/Fable, Gemini, Kimi, GLM, and DeepSeek-V4, plus Claude model selection and Astra Skills/AGENTS.md audits.
★not rated 6
changed 19d agoA
tokens not measured
originalMIT
LEAP: LLM Engineered Application Pattern. A philosophy where code is a commodity — only prompts, schemas, and tests are source of truth. AI generates all application code on demand.
★not rated 6 6mo agoA
tokens not measured
originalMIT
Finds the named technique your problem needs and folds its phrasing into the request before the work starts. Hundreds of named techniques, ranked locally, no network.
★not rated 6 9d agoA
tokens not measured
originalMIT
Interactive prompt structuring skill using a 7-component framework derived from Anthropic's published prompting guide. Tier-adaptive dialogue, deterministic component rubric, and activation-regression fixtures.
A quiet prompting coach. Scores how you ask, names one thing to sharpen, and remembers your recurring patterns across sessions so the feedback compounds. Works for anyone using AI — not tied to any one domain.
★not rated 5 3mo agoA
tokens not measured
originalMIT
Rescues messy, typo-ridden, or half-formed prompts so Claude understands you the first time. Confirms its reading with a one-line '✓ understood · …' and keeps reviewable history and stats.
★not rated 5 4mo agoA
tokens not measured
originalMIT
Skills for Adaline platform integration: logging traces and spans, managing prompts, datasets, evaluators, evaluations, deployments, and AI providers via API and SDKs.
Build high-quality, structured prompts using the 10-block framework from Anthropic's 'Prompting 101' workshop. Interview-style: asks 4 targeted questions, picks a task-type template, and produces a production-grade prompt with annotations and a test plan.
★not rated 4 5mo agoA
tokens not measured
originalMIT
Cleans up rough prompts before the assistant acts on them: fixes typos, expands terse asks, repairs broken {{template}} placeholders, and strips filler openers. Includes a /fix-prompt command, an auto-invokable skill, and a non-blocking UserPromptSubmit nudge.
Turns rough, vague prompts into precise, context-enriched prompts. Explores the current codebase read-only, weaves real file paths and symbols into the rewritten prompt, and never implements the task itself.
★not rated 3 1mo agoA
tokens not measured
originalMIT