What is agent-plugins CLAUDE.md?

A repository guide for a marketplace containing plugins for Claude Code, an AI coding assistant. The plugins cover Zig guidance, test-driven development (writing tests before implementation), code review, pair programming, knowledge management, and coordinating multiple agents.

Use it when adding or maintaining these Claude Code plugins, their instructions, marketplace metadata, or the agent-fleet and review workflows described in the guide.

It gives agents the repository structure and the intended purpose of each plugin, so changes follow the project’s conventions instead of treating the marketplace as an ordinary application.

agent-plugins CLAUDE.md — Instructions file by kelp

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 instructions/kelp/agent-plugins/claude-md

Clone the repo

git clone --depth 1 https://github.com/kelp/agent-plugins

Per session 1,885 This file is loaded in full into every session.

When invoked 1,885 The same file — it is already loaded in full.

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.

Model	Per session	Once invoked
Fable 5	$0.01885	$0.01885
Opus 5	$0.00942	$0.00942
Sonnet 5	$0.00377	$0.00377
Haiku 4.5	$0.00188	$0.00188

Measured 2d ago against content hash 91b6438b7e81, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-plugins CLAUDE.md 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.md · 204 lines

How it starts

The opening of the file, as written. The whole thing — 204 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

What This Repo Is

A Claude Code plugin marketplace containing these plugins:

zig-claude-kit -- corrective context for Zig that fixes Claude's outdated training data; auto-detects 0.15.x vs 0.16.x and uses the matching reference
tiger-style -- TigerBeetle's Tiger Style methodology for Zig (assertions, bounded loops, static memory, snake_case naming, 70-line / 100-col limits)
tdd-pipeline -- language-agnostic TDD pipeline: four agent roles across a seven-stage pipeline (the red/verify/integrate gates run in the orchestrator, not as separate agents)
cross-review -- multi-model code review with cross-validation using Claude and GPT-5.5
codex-pair -- pair programming with a persistent Codex partner over one long-lived thread (model set by the user's Codex install), with a Fable judge for deadlocks
knowledge-forge -- cross-session routing and capture for a three-layer personal knowledge base
fleet-efficiency -- agent-fleet rules and procedure: context-handoff, prompt-cache, and model-tier rules for fanning out many subagents, plus the two-lead operating procedure above the fan-out
next-todo -- land one TODO.md PR slice: plan review to consensus, red-green TDD, draft PR, comment drain; project bindings live in CLAUDE.md

Users install via /plugin marketplace add kelp/agent-plugins, then /plugin install <name>@agent-plugins.

Repository Structure

.claude-plugin/marketplace.json    # plugin registry
plugins/
  zig-claude-kit/
    .claude-plugin/plugin.json     # manifest (version here)
    skills/                        # zig-init, zig-patterns, zig-check
    hooks/hooks.json               # SessionStart hook
    scripts/                       # eval suite, session-start,
                                   #   detect-zig-version
    docs/0.15/                     # fragment, breaking changes ref
    docs/0.16/                     # fragment, breaking changes ref
  tiger-style/
    .claude-plugin/plugin.json     # manifest (version here)
    skills/                        # tiger-init, tiger-patterns, tiger-check
    hooks/hooks.json               # SessionStart hook
    scripts/                       # session-start
    docs/                          # fragment, reference doc
  tdd-pipeline/
    .claude-plugin/plugin.json     # manifest (version here)
    skills/                        # tdd-orchestrate, tdd-init
    agents/                        # test-writer, test-reviewer,
                                   #   implementer, code-reviewer
    docs/                          # fragment, methodology ref
  cross-review/
    .claude-plugin/plugin.json     # manifest (version here)
    skills/cross-review/           # orchestrator skill
    agents/                        # reviewer, validator
    docs/                          # fragment, codex prompts,
                                   #   merge/output reference
  codex-pair/
    .claude-plugin/plugin.json     # manifest (version here)
    skills/pair/                   # orchestrator skill
    scripts/                       # codex-pair.mjs, lib.mjs
    tests/                         # cli.test.mjs, lib.test.mjs
  knowledge-forge/
    .claude-plugin/plugin.json     # manifest (version here)
    skills/                        # kb-capture, kb-ingest,
                                   #   kb-research-policy
    docs/                          # fragment
  fleet-efficiency/
    .claude-plugin/plugin.json     # manifest (version here)
    skills/                        # fleet-efficiency (fan-out
                                   #   rules), fleet-lead (two-lead
                                   #   procedure); both auto-discovered
  next-todo/
    .claude-plugin/plugin.json     # manifest (version here)
    skills/                        # next-todo, next-todo-init
    docs/                          # CLAUDE.md binding fragment

Read the full file on GitHub · 204 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.

2d ago First seen · 204 lines · 1,885 tokens per session scan A 91b6438b7e81

Subscribe to this mod's changes

agent-plugins CLAUDE.md is an instructions file published in the GitHub repository kelp/agent-plugins (2 stars, last pushed 9d ago), licensed MIT. It adds 1,885 tokens to every session, about $0.0094 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 2026-08-31.

claude claude-code claude-code-plugins llm-training zig zig-lang

agent-plugins CLAUDE.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

CLAUDE.md

What This Repo Is

Repository Structure

What this file has done since we first saw it

Other instructions, from other repositories

agentops AGENTS.md

claude-plugins CLAUDE.md

cortex CLAUDE.md

rtailscale GEMINI.md

rgotify AGENTS.md

rapprise AGENTS.md