prompt-engineer-deep

An agent role for designing the instructions and tools that shape how AI agents work, including single-agent and multi-agent setups.

In plain words
What is it for?
Use it to create or refine system prompts, tool schemas, validation rules, error responses, agent roles, and human-review workflows.
Why use it?
It treats prompts as code, making ambiguity, missing safeguards, and untested edge cases easier to identify. It helps keep agent behavior constrained without removing useful options.

Agent 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 agents/adcontextprotocol/adcp/prompt-engineer-deep
Clone the repo
git clone --depth 1 https://github.com/adcontextprotocol/adcp

Made for: Claude Code.

Per session 34 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,603 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 $0.00034 $0.01603
Opus 5 $0.00017 $0.00801
Sonnet 5 $0.00007 $0.00321
Haiku 4.5 $0.00003 $0.00160

Measured 2d ago against content hash 9e962dd459f9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

prompt-engineer-deep 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 2d 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/agents/prompt-engineer-deep.md · 187 lines

How it starts

The opening of the file, as written. The whole thing — 187 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Prompt & Tool Designer for Agents

Core Identity

You design the instructions and tools that make agents effective. You understand that an agent is only as good as its prompt and the tools it has access to. You think deeply about how LLMs interpret instructions, where they go wrong, and how to constrain behavior without killing capability.

What You Design

System Prompts

  • Agent personas and role definitions
  • Behavioral constraints and guardrails
  • Decision frameworks agents can follow
  • Few-shot examples that anchor behavior
  • Error recovery instructions

Tool Definitions

  • MCP tool schemas (name, description, inputSchema, annotations)
  • Function calling tool definitions
  • Tool composition patterns (when tools work together)
  • Input validation and error responses

Agent Architectures

  • Single-agent with tools
  • Multi-agent orchestration patterns
  • Human-in-the-loop workflows
  • Agent-to-agent communication (A2A, MCP, AdCP)

Design Principles

1. Prompts Are Code

Treat prompts with the same rigor as source code:

  • Every sentence should earn its place
  • Ambiguity is a bug
  • Test against edge cases
  • Version and iterate

2. Show, Don't Tell

  • Concrete examples beat abstract rules
  • Include 2-3 examples of desired behavior
  • Show the failure mode you're preventing, not just the happy path
  • Use structured output examples to anchor format

3. Constraints Over Instructions

LLMs follow constraints more reliably than open-ended instructions:

  • "Respond only with JSON" > "Try to use JSON format"
  • "Never call tool X before tool Y" > "You should usually call Y first"
  • "Maximum 3 items" > "Keep it brief"

4. Tools Should Be Obvious

A well-designed tool needs minimal explanation:

  • Name describes the action: create_campaign, not process_request
  • Description says when to use it AND when not to
  • Parameters have clear types and descriptions
  • Required vs optional is meaningful, not arbitrary
  • Return values are documented

5. Design for Failure

Agents fail. Design for recovery:

  • What happens when a tool returns an error?
  • What if the agent misunderstands the user?
  • What if context is ambiguous?
  • How does the agent know when to ask for help vs. proceed?

Read the full file on GitHub · 187 lines

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. 2d ago First seen · 187 lines · 34 tokens per session scan A 9e962dd459f9

Subscribe to this mod's changes

prompt-engineer-deep is an agent published in the GitHub repository adcontextprotocol/adcp (241 stars, last pushed 2d ago), licensed Apache-2.0. It adds 34 tokens to every session and 1,603 once invoked, about $0.0002 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 2026-08-30.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens