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 instructions/igniteui/igniteui-mcp-testbed/claude-mdgit clone --depth 1 https://github.com/IgniteUI/igniteui-mcp-testbedWrote 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.
[](https://agentmods.dev/instructions/igniteui/igniteui-mcp-testbed/claude-md)<a href="https://agentmods.dev/instructions/igniteui/igniteui-mcp-testbed/claude-md"><img src="https://agentmods.dev/badge/instructions/igniteui/igniteui-mcp-testbed/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 5.1 | $0.16891 | $0.16891 |
| Opus 5 | $0.08446 | $0.08446 |
| Sonnet 5 | $0.03378 | $0.03378 |
| Haiku 4.5 | $0.01689 | $0.01689 |
Grade A, and why
igniteui-mcp-testbed 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 6d 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 — 261 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 is
A single-container appliance for exercising the Ignite UI AI toolchain (the Ignite UI CLI MCP server, the Theming MCP server, and the Agent Skills) against opencode. A small Express web wizard collects the user's choices, scaffolds an Ignite UI project, wires the AI config, and hands off to the opencode web UI. Each session runs in a fresh, ephemeral rootless Podman container.
This repo's logic (the MCP translation and server classification) is unit-/syntax-tested, but the README notes it has not been run end-to-end against real igniteui-cli / opencode / Podman — treat first builds as a shakedown, and expect the "adjust to your packages" spots (see below) to need tuning.
Commands
./run.sh build # podman build -t localhost/igniteui-testbed:latest .
./run.sh build --prune # build, then `podman image prune -f` to drop dangling <none> images
./run.sh # run a fresh ephemeral container; publishes ports 8080 / 4096 / 5000
./run.sh --matrix-config <file> # run with a matrix JSON config: bind-mounts it to
# /matrix-config.json, sets MATRIX_CONFIG, and (by default)
# auto-runs the matrix headlessly; the UI prefills from it
./stop.sh [<sess>] # stop running testbed container(s) (all, or one by session id)
# all four scripts take -h/--help/help (PowerShell: -Help, `help`, -? or Get-Help) and print usage
npm start # run the wizard backend directly (node src/server.ts), for host-side dev
npm run typecheck # tsc --noEmit gate over backend (tsconfig.json) + frontend (web/tsconfig.json)
npm run build:web # esbuild web/main.ts -> public/vendor/app.js (run alongside `npm start` for host dev)
npm run dev:web # build:web in --watch mode
On Windows, run.ps1 / stop.ps1 are PowerShell ports of run.sh / stop.sh with the
same arguments (.\run.ps1 build, .\run.ps1 build -Prune, .\run.ps1,
.\run.ps1 -MatrixConfig <file> [-Validate], .\stop.ps1 [<sess>]). The prune step (--prune /
-Prune) only runs after a successful build and removes only untagged images. Keep the
two implementations in sync — a change to one platform's run/stop logic should be
mirrored in the other.
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.
- 6d ago First seen · 261 lines · 16,891 tokens per session scan A 12249066b207
igniteui-mcp-testbed CLAUDE.md is an instructions file published in the GitHub repository IgniteUI/igniteui-mcp-testbed (50 stars, last pushed 9d ago), licensed MIT. It adds 16,891 tokens to every session, about $0.0845 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 instructions, from other repositories
appwrite AGENTS.md
AGENTS.md instructions for appwrite/appwrite, covering appwrite, commands, stack, layout and libraries.
selfhost-ai CLAUDE.md
Claude Code instructions for kossakovsky/selfhost-ai, covering claude.md, project overview, core architecture, key files and installation flow.
openlakeforge AGENTS.md
AGENTS.md instructions for malon64/openlakeforge, covering agent and contributor guide, what this project is, non-goals, orientation — read in this order and repository map.
docker-zoneminder CLAUDE.md
Instructions for jantman/docker-zoneminder, covering claude.md, project overview, build and test, ci/cd and architecture.
mitos CLAUDE.md
Instructions for mitos-run/mitos, covering claude.md, project overview, operating principles, commands and architecture.
MoaV AGENTS.md
AGENTS.md instructions for MotherofallVPNs/MoaV, covering what moav is, install & run, everyday operations, dashboards & access — what to hand the owner and where things live.