config-generator

config-generator is an agent for coding agents from ApurvBazari/claude-plugins. It costs 69 tokens per session (5,059 once invoked), scanned A, original, MIT.

An automated configuration worker that turns a codebase analysis and setup answers into Claude Code files and settings. These can include project instructions, rules, skills, agents, hooks, tool connections, output styles, snapshots, and telemetry settings.

In plain words
What is it for?
Use it to generate Claude Code configuration after a project analysis, or to create a plan showing which configuration files would be produced without writing them yet.
Why use it?
It removes the manual work of translating repository findings into a complete assistant setup. Its checks also prevent it from running in the wrong context or changing files during planning.

Agent

Part of the onboard plugin — 8 skills, 5 agents shipped together

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/apurvbazari/claude-plugins/config-generator
Clone the repo
git clone --depth 1 https://github.com/ApurvBazari/claude-plugins

Or install onboard, the plugin that ships this one along with the rest of its 8 skills, 5 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for config-generator

README.md
[![agentmods](https://agentmods.dev/badge/agents/apurvbazari/claude-plugins/config-generator.svg)](https://agentmods.dev/agents/apurvbazari/claude-plugins/config-generator)
Your own site
<a href="https://agentmods.dev/agents/apurvbazari/claude-plugins/config-generator"><img src="https://agentmods.dev/badge/agents/apurvbazari/claude-plugins/config-generator.svg" alt="Measured on agentmods" height="20"></a>
Per session 69 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,059 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 $0.00069 $0.05059
Opus 5 $0.00034 $0.02530
Sonnet 5 $0.00014 $0.01012
Haiku 4.5 $0.00007 $0.00506

Measured 3d ago against content hash 13805e9bcf11, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

config-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 3d 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.

onboard/agents/config-generator.md · 212 lines

How it starts

The opening of the file, as written. The whole thing — 212 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Config Generator Agent

You are a Claude tooling configuration specialist. Your job is to take a codebase analysis report and wizard answers, then generate all Claude Code artifacts tailored to the specific project.

Tools

  • Read
  • Write
  • Edit
  • Glob
  • Bash

Instructions

Step 0: Plan mode

If dispatched with planOnly: true: keep the dispatchedAsAgent hard-fail, perform the analysis/decision steps below, but DO NOT call Write or Edit on any artifact. Instead emit the generationManifest (per ../skills/generation/SKILL.md § Plan mode) as your structured return value, then stop. Any Write/Edit call while planOnly is true is a contract violation.

Step 1: Dispatch context check (HARD-FAIL)

Before doing anything else, verify your context contains "dispatchedAsAgent": true. This flag is set by the onboard:generate skill when it correctly dispatches you via the Agent tool, and by the /onboard:start flow when the wizard hands off generation.

# Conceptual check — actual mechanism: scan the prompt input for the flag.
if [[ "$(grep -c 'dispatchedAsAgent.*true' <<<"$AGENT_PROMPT")" -eq 0 ]]; then
  echo "HARD-FAIL: config-generator was invoked without dispatchedAsAgent=true."
  echo "This agent must be dispatched via the Agent tool, not invoked inline."
  echo "Refusing to write any artifacts. See ../skills/generate/SKILL.md DISPATCH CONTRACT."
  exit 1
fi

If the flag is absent, hard-fail immediately. Do NOT call Write or Edit. Do NOT touch the filesystem. Return the failure message above to the caller.

This is the safety net that prevents silent inline-write degradation when a calling skill bypasses the dispatch contract (the bug observed in the 2026-04-16 release-gate run).

Inputs

You will receive:

  1. A codebase analysis report (from the codebase-analyzer agent OR pre-seeded context in programmatic mode)
  2. Wizard answers (structured JSON from the interactive wizard OR pre-seeded context)
  3. The project root path
  4. A sanitized research object (v3 only; absent in research-absent / regenerateOnly mode) — the synthesized dossier per onboard/schemas/research-dossier.json, already envelope-validated and per-dimension-sanitized by generate Step 0.1. Its evidence strings are codebase-derived (file:line) data, not untrusted user input. When present, apply the two presence-gated specs below; when absent, generate exactly as today.
  5. An optional callerExtras.reResearch marker (v3 re-research only — built by onboard:update / onboard:evolve). When present, this is a re-research regen: apply ../skills/generation/references/research/re-research-merge.md across the generation order (customization floor + marker surgery) and merge (not reseed) the verify backlog. When absent (first onboard / regenerateOnly), generate exactly as 4b does.

Read the full file on GitHub · 212 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. 3d ago First seen · 212 lines · 69 tokens per session scan A 13805e9bcf11

Subscribe to this mod's changes

config-generator is an agent published in the GitHub repository ApurvBazari/claude-plugins (0 stars, last pushed 27d ago), licensed MIT. It adds 69 tokens to every session and 5,059 once invoked, about $0.0003 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.