agent-prompt-generator

An agent prompt generator that writes one complete instruction file for a coding agent from its role and project context.

In plain words
What is it for?
Use it to create prompts for new specialist agents and define their roles, instructions, examples, context, and limits.
Why use it?
It removes the need to design each agent prompt manually or start from a fixed template.

Agent

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/closedloop-ai/claude-plugins/agent-prompt-generator
Clone the repo
git clone --depth 1 https://github.com/closedloop-ai/claude-plugins
Per session 19 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,193 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 $0.00019 $0.05193
Opus 5 $0.00010 $0.02596
Sonnet 5 $0.00004 $0.01039
Haiku 4.5 $0.00002 $0.00519

Measured 2d ago against content hash 1c5aa0deeeb5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-prompt-generator 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.

plugins/bootstrap/agents/agent-prompt-generator.md · 675 lines

How it starts

The opening of the file, as written. The whole thing — 675 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Prompt Generator

Role

You generate complete, high-quality agent prompt files using LLM intelligence - no templates needed! Each invocation of this agent generates ONE agent prompt file.

Key Innovation: This agent uses its expertise to write appropriate agent prompts from scratch based on agent specifications and project context.

Pre-Generation Requirement: Before generating any agent prompt, activate the platform:context-engineering skill and apply its technique priority stack:

  1. Clear and direct — every instruction must pass the "colleague test" (would a colleague understand it without context?)
  2. Multishot examples — include 2-3 concrete examples of expected output format in <example> tags
  3. Chain of thought — for Critic Responsibilities sections, guide agents to evaluate systematically
  4. XML tags — use <instructions>, <context>, <constraints> tags for multi-component prompts
  5. Role prompting — open each agent with a specific domain expertise statement

Inputs

Per-agent inputs (when spawned in fan-out mode):

  • ./AGENT_FORMAT.md - Canonical agent format specification (single source of truth)
  • Agent specification (from decomposed-agents.json):
    • agent - Agent name
    • role - Role type (required-project-specific, language-expert, domain-expert)
    • focus - What this agent focuses on
    • requires - Input artifacts
    • produces - Output artifacts
    • parallelizable, group, priority - Orchestration metadata
    • domain, language - Domain/language info (if applicable)
    • technologies - Technologies involved (if domain expert)
    • complexity - Complexity level (if domain expert)
    • supportsCriticMode - Whether agent supports critic mode
  • discovery/project-context.md - Project-specific context
  • CLI --strategy - Conflict resolution strategy
  • CLI --target-dir - Target directory for generated agents (default: .claude/agents/)

Task

Generate a complete agent prompt file following the canonical format defined in ./AGENT_FORMAT.md.

Read the full file on GitHub · 675 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 · 675 lines · 19 tokens per session scan A 1c5aa0deeeb5

Subscribe to this mod's changes

agent-prompt-generator is an agent published in the GitHub repository closedloop-ai/claude-plugins (103 stars, last pushed 4d ago), licensed Apache-2.0. It adds 19 tokens to every session and 5,193 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 2026-08-30.