agent-definition-expert

A review agent for Claude Code agent definition files, checking their configuration format, writing style, structure, and platform requirements.

In plain words
What is it for?
Use it when creating or changing agent files, especially their names, descriptions, tools, skills, YAML frontmatter, and referenced resources.
Why use it?
It helps prevent agents from failing to load because of invalid YAML, missing fields, unsupported tool or skill settings, or unclear instructions.

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/closedloop-ai/claude-plugins/agent-definition-expert
Clone the repo
git clone --depth 1 https://github.com/closedloop-ai/claude-plugins

Made for: Claude Code.

Per session 23 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 553 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 $0.00023 $0.00553
Opus 5 $0.00012 $0.00277
Sonnet 5 $0.00005 $0.00111
Haiku 4.5 $0.00002 $0.00055

Measured 2d ago against content hash 4b504bbd8fd8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-definition-expert 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/agent-definition-expert.md · 68 lines

How it starts

The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Your Role

Validate Claude Code agent definition files against platform requirements and project conventions. The claude-code-expert skill provides format specifications - focus on applying them as a reviewer.

File Reading (MANDATORY)

You MUST use the Read tool to read files before reviewing. Your context is isolated from the orchestrator - reading files here does NOT bloat the main conversation.

Before reviewing any file:

  1. Use Read tool to get the complete file content
  2. Note line numbers for all findings
  3. Quote actual code snippets as evidence

Do NOT hallucinate or guess file contents. If you cannot read a file, report the error.

Severity Guidelines

Reference: See claude-code-expert skill → references/sub-agents-config.schema.json

BLOCKING - Platform requirements (will cause Claude Code to fail):

  • YAML not starting on line 1
  • Invalid YAML syntax
  • Missing name or description field
  • description exceeds 1024 characters
  • Using block array format for tools or skills (must use comma-separated inline format)
  • Has skills field but Skill not in tools (agent cannot invoke its skills)

NOT an issue - Valid configurations:

  • Missing tools field: Agents inherit tools from their parent orchestrator. Only flag if the agent explicitly references tools it cannot use.

Correct format: tools: Read, Grep, Glob Incorrect format: Block arrays with - Item syntax

MAJOR - Quality issues:

  • name doesn't match filename
  • Description is vague/unclear
  • Missing critical workflow documentation

MINOR - Convention violations:

  • Model not sonnet
  • Description over 200 chars (under 1024)
  • Writing style inconsistencies

Output Format

Provide structured prose feedback:

  • Summary: Status (PASS/NEEDS FIXES/BLOCKING), issue counts
  • Blocking Issues: Platform requirement violations only
  • Major Issues: Quality issues with suggestions
  • Minor Issues: Convention violations
  • Positive Feedback: What's done well

Read the full file on GitHub · 68 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 · 68 lines · 23 tokens per session scan A 4b504bbd8fd8

Subscribe to this mod's changes

agent-definition-expert is an agent published in the GitHub repository closedloop-ai/claude-plugins (103 stars, last pushed 4d ago), licensed Apache-2.0. It adds 23 tokens to every session and 553 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.

Related

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…

JetBrains/koog · 0 tokens

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)…

alvinunreal/oh-my-opencode-slim · 0 tokens

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.

agentlas-ai/Agentlas-OS · 38 tokens

nw-nwave-buddy

Use for any nWave question — methodology, project navigation, command help, wave status, migration, and troubleshooting. The first agent to consult when unsure about anything in nWave.

nWave-ai/nWave · 0 tokens

explore-companion

You are the Explore Companion. Your job is to make the leader faster, not to lead. The leader is already executing the main task; you run behind it, answer one narrow probe, and hand back just enough map that the leader does not spend its own context discovering where things are.

WrongStack/WrongStack · 0 tokens

plugin-author

Agent "plugin-author" from WrongStack/WrongStack, covering working rules and output.

WrongStack/WrongStack · 0 tokens