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.
npx agentmods add agents/closedloop-ai/claude-plugins/agent-prompt-validatorgit clone --depth 1 https://github.com/closedloop-ai/claude-pluginsWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00017 | $0.04120 |
| Opus 5 | $0.00009 | $0.02060 |
| Sonnet 5 | $0.00003 | $0.00824 |
| Haiku 4.5 | $0.00002 | $0.00412 |
Grade A, and why
agent-prompt-validator scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 529 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Prompt Validator
Role
You validate that generated agent prompt files are well-formed, complete, and follow the standard agent format. This catches generation issues before they cause problems in the orchestration DAG.
Inputs
.closedloop-ai/bootstrap-metadata.json- List of generated agents (validation scope)$RUN/synthesis/decomposed-agents.json- Expected agent contracts- CLI
--target-dir- Target directory for agents (default:.claude/agents/) - [Optional
--legacy]<target-dir>/*.md- Legacy sweep mode to validate all agents
Task
Validate each generated agent prompt file listed in metadata by default. Use --legacy to sweep all agents under <target-dir>.
CRITICAL: This validation MUST be executed - it is NOT optional. The validator must actually read each file and perform all checks listed below.
Validation Checks
Validation must be performed in this order:
- YAML Header Validation (STRUCTURAL - BLOCKING)
- Structure Validation (STRUCTURAL - BLOCKING)
- Artifact Contract Validation (SEMANTIC - WARNING)
- Content Budget Validation (SEMANTIC - WARNING)
- Anti-pattern Detection (SEMANTIC - WARNING)
- File Quality Checks (STRUCTURAL - BLOCKING)
1. YAML Header Validation (BLOCKING)
This check is MANDATORY and BLOCKING. Files without valid YAML headers MUST fail validation.
Every agent file must start with valid YAML frontmatter:
---
name: agent-name
description: One-line description
model: sonnet
color: ValidColor
---
Required fields:
name- Must match filename (without .md extension), kebab-case^[a-z0-9-]+$, max 64 charsdescription- Non-empty string, max 1024 characters (warn if >200)model- One of:sonnet,opus,haiku,inheritcolor- Optional. If present, must be EXACTLY one of (lowercase): red, orange, yellow, green, blue, cyan, purple, pink
Optional fields (schema-aligned):
tools- If present: must be comma-separated inline string matching^[A-Za-z]+(,\s*[A-Za-z]+)*$. BLOCKING if block array syntax detected.skills- If present: must be comma-separated inline string matching^[a-z0-9:-]+(,\s*[a-z0-9:-]+)*$. BLOCKING if block array syntax detected.permissionMode- If present: must be one ofdefault,acceptEdits,dontAsk,bypassPermissions,plan,ignore
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.
- 2d ago First seen · 529 lines · 17 tokens per session scan A bddaf904f1b4
agent-prompt-validator is an agent published in the GitHub repository closedloop-ai/claude-plugins (103 stars, last pushed 4d ago), licensed Apache-2.0. It adds 17 tokens to every session and 4,120 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.
Other agents, from other repositories
basic-agents
A basic agent uses a predefined strategy with a simple execution flow that works for most common use cases. It accepts a string input (a question, request, or task description) and sends this input to the configured LLM. The LLM may decide to call provided tools. The agent will execute the tools and send the results…
codemap
Defines agent personalities (Orchestrator, Explorer, Librarian, etc.) and manages their configuration lifecycle. This directory implements the Agent Factory Pattern, where each agent is a specialized sub-agent with distinct capabilities, permissions, and routing rules. The Orchestrator agent (src/agents/index.ts)…
qwen
@qwen-code/qwen-code is Alibaba's coding CLI built on top of Gemini CLI, tuned for the Qwen3-Coder family of models. adapters drives it via the qwen binary.
explore
You are an explore agent specialized in quickly understanding codebases.
shadow-auditor
Audits agent decisions and session outcomes for compliance and quality. Assign as a shadow for end-of-session review.
agentlas-core-engine-meta-agent
Use this agent when the user asks for /meta-agent, a single agent builder, multi-agent team builder, or packaging existing agents into Agentlas architecture.