agentleFS
viasrijanviasrijan/crazy-skills/skills/productivity/prompt-enhancement/SKILL.md

prompt-enhancement

Skill0 starsChanged 3 months agoMITView on GitHub

Use when enhancing or optimizing input prompts — transform vague prompts into precision instructions with RSCIT, hallucination guards, and compression to get the best results.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Prompt Enhancement

Transform weak or vague prompts into precision-engineered instructions. Composite of brixtonpham/claude-config prompt-enhancer + llm-prompt-optimizer RSCIT.

When to use

  • Input is vague, inconsistent, or hallucination-prone
  • Need structured/JSON output reliably
  • Want to cut input tokens without losing effectiveness
  • This skill is always-on in Crazy Skills — also wired as UserPromptSubmit hook for auto-enhancement

Workflow

1. Diagnose

ProblemSymptomFix
Too vagueGeneric answersAdd Role + Context + Constraints
No structureUnformattedSpecify output format
HallucinationConfident wrongAdd "say I don't know if unsure"
InconsistentDifferent each runAdd few-shot examples
Too longVerboseAdd length constraints

2. Apply RSCIT

  • R — Role: Who is the AI? (e.g., "senior ML engineer")
  • S — Situation: What context? (e.g., "1 year Python, no ML background")
  • C — Constraints: Rules and limits (e.g., "≤200 words, no formulas, use analogy")
  • I — Instructions: What exactly to do?
  • T — Template: What output looks like?

Before: Explain machine learning. After: You are a senior ML engineer. Context: junior with 1 year Python. Task: explain supervised ML simply. Constraints: analogy, ≤200 words, no formulas, one next step. Format: plain prose.

3. Compress filler

Please carefully analyze the following code and provide... → Analyze this code: explain what it does, how it works, flag issues.

4. Hallucination guard

Answer based ONLY on provided context. If not contained, respond exactly: "I don't have enough information."

Verification

  • Role, Situation, Constraints, Instructions, Template are explicit
  • Output format is explicitly defined (JSON/markdown/bullets)
  • Hallucination guard present for factual tasks
  • Prompt is compressed (filler removed) without losing task-critical instructions

References

  • Source: brixtonpham/claude-config prompt-enhancer, mohamednaeem92-max/OPENCODE-6-2026 llm-prompt-optimizer
  • Hook: UserPromptSubmit → prompt-enhancement (optional auto-mode in opencode.json)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.