mep

mep is a command for Claude Code from BACH-AI-Tools/pdf-reader-mcp. It costs 11 tokens per session (450 once invoked), scanned A, original, MIT.

A command that rewrites a verbose request into a Minimal Effective Prompt, meaning a shorter prompt that keeps only information the AI cannot infer. It focuses on requirements, constraints, domain knowledge, outcomes, and acceptance criteria.

In plain words
What is it for?
Use it to simplify prompts, separate required business rules from implementation details, and state the desired behavior more clearly.
Why use it?
It removes repeated technical or project details that an AI assistant can discover or already understands. This makes the request shorter while retaining the decisions that affect the result.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/bach-ai-tools/pdf-reader-mcp/mep
Clone the repo
git clone --depth 1 https://github.com/BACH-AI-Tools/pdf-reader-mcp

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for mep

README.md
[![agentmods](https://agentmods.dev/badge/commands/bach-ai-tools/pdf-reader-mcp/mep.svg)](https://agentmods.dev/commands/bach-ai-tools/pdf-reader-mcp/mep)
Your own site
<a href="https://agentmods.dev/commands/bach-ai-tools/pdf-reader-mcp/mep"><img src="https://agentmods.dev/badge/commands/bach-ai-tools/pdf-reader-mcp/mep.svg" alt="Measured on agentmods" height="20"></a>
Per session 11 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 450 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00011 $0.00450
Opus 5 $0.00005 $0.00225
Sonnet 5 $0.00002 $0.00090
Haiku 4.5 $0.00001 $0.00045

Measured 3d ago against content hash e2ddf8364142, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

mep scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/mep.md · 72 lines

What it actually says

MEP - Minimal Effective Prompt

Context

User's original prompt:

$ARGUMENTS

Your Task

Analyze the user's prompt above and refactor it into a Minimal Effective Prompt (MEP) that:

Remove Unnecessary Context

❌ Remove information that AI already knows:

  • Current date/time (AI has access via hooks)
  • System information (platform, CPU, memory - provided automatically)
  • Project structure (AI can search codebase)
  • Tech stack (AI can detect from package.json and code)
  • File locations (AI can search)
  • Existing code patterns (AI can search codebase)

Keep Essential Information

✅ Keep only what AI cannot infer:

  • Specific business requirements
  • User preferences or constraints
  • Domain-specific knowledge
  • Desired outcome or behavior
  • Acceptance criteria

Apply MEP Principles

  1. Be Specific About What, Not How

    • ❌ "Create a React component with useState hook, useEffect for data fetching, proper error handling..."
    • ✅ "Add user profile page with real-time data"
  2. Trust AI's Knowledge

    • ❌ "Using TypeScript with proper types, following our code style..."
    • ✅ "Add user authentication" (AI will use TypeScript, follow existing patterns)
  3. Focus on Intent

    • ❌ "I need a function that takes an array and returns unique values using Set..."
    • ✅ "Remove duplicate items from the list"
  4. Remove Redundancy

    • ❌ "Add comprehensive error handling with try-catch blocks and proper error messages..."
    • ✅ "Add error handling" (comprehensive is default)

Output Format

Provide the refactored MEP prompt as a single, concise statement (1-3 sentences max) that captures the essence of the user's intent.

Original: [quote the original]

MEP Version: [your refactored minimal prompt]

Removed Context: [list what was removed and why - explain that AI already has this info]

Preserved Intent: [confirm the core requirement is maintained]

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 72 lines · 11 tokens per session scan A e2ddf8364142

Subscribe to this mod's changes

mep is a command published in the GitHub repository BACH-AI-Tools/pdf-reader-mcp (0 stars, last pushed 6mo ago), licensed MIT. It adds 11 tokens to every session and 450 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-01.