Irrlicht: Skill for Claude Code

.claude/skills/ir:onboarding-factory/record/SKILL.md

ir:onboarding-factory/record is a skill for Claude Code from ingo-eichhorst/Irrlicht. It costs 106 tokens per session (6,588 once invoked), scanned D, original, MIT.

A workflow for recording and verifying one test scenario with a live coding agent. It checks prerequisites, runs the agent through a recording service, verifies captured events, refreshes the replay reference, and commits the result.

In plain words
What is it for?
It is for handling recordable onboarding scenarios, including preparing missing driver steps, checking authentication or local services, validating websocket observations, and updating committed replay data.
Why use it?
It reduces the chance that a recording is incomplete, made under missing conditions, or inconsistent with the expected replay data.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: reads .claude/ paths; mentions subagents; reads ~/.codex paths.

This is ingo-eichhorst/Irrlicht's own configuration. It tells Claude Code how to work on Irrlicht itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Irrlicht configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is bash tools/onboarding-factory/scripts/lib/completeness-check.sh <staging-dir>.

Reuse

Borrowing it

Nothing to install: this file belongs to ingo-eichhorst/Irrlicht. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/ingo-eichhorst/Irrlicht/main/.claude/skills/ir:onboarding-factory/record/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/ingo-eichhorst/Irrlicht

Made for: Claude Code.

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 ir:onboarding-factory/record

README.md
[![agentmods](https://agentmods.dev/badge/skills/ingo-eichhorst/irrlicht/record/github.svg)](https://agentmods.dev/skills/ingo-eichhorst/irrlicht/record)
Your own site
<a href="https://agentmods.dev/skills/ingo-eichhorst/irrlicht/record"><img src="https://agentmods.dev/badge/skills/ingo-eichhorst/irrlicht/record/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ir:onboarding-factory/record

Your own site · 80×15
<a href="https://agentmods.dev/skills/ingo-eichhorst/irrlicht/record"><img src="https://agentmods.dev/badge/skills/ingo-eichhorst/irrlicht/record.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 106 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 6,588 The whole file, excluding the scripts and references it only reads on demand.
Security scan D 2 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 2 findings, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high Agent Snooping · line 94
    Skill reads from agent configuration directories (.claude/, .codex/, .gemini/). These directories may contain API keys, personal settings, and other credentials that the skill has no legitimate need to access.
    Fix: Remove all code or instructions that access agent configuration directories (.claude/, .codex/, .gemini/). If configuration values are needed, pass them explicitly as parameters or environment variabl
  • medium Rogue Agent · line 93
    Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
    Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
How audits are shown
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 5.1 $0.00106 $0.06588
Opus 5.5 $0.00042 $0.02635
Sonnet 5.5 $0.00021 $0.01318
Haiku 4.5 $0.00011 $0.00659

Measured 1mo ago against content hash be3a5bd95fc7, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade D, and why

ir:onboarding-factory/record scanned grade D with 2 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 1mo 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

(`~/.claude/settings.json`, `~/.codex/hooks.json`, …), because a dev daemon

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

half-promoted directory to `rm -rf` before re-promoting). Either tighten the
.claude/skills/ir:onboarding-factory/record/SKILL.md · 502 lines

How it starts

The opening of the file, as written. The whole thing — 502 lines — stays where its author put it; the contents beside it link to each section on GitHub.

record

You run as a focused subagent with no parent context. This verb DRIVES A LIVE AGENT CLI and SPENDS API TOKENS — auth is set up out-of-band, so run it without ceremony (no key checks, no "this will spend money" prompts). It must be serialized: only one record runs against the live daemon at a time. When done, return only the "Return contract" block.

Preconditions

  1. The cell is assessed and on a recordable route. Read it:

    of status --agent <agent> --scenario <scenario> --json
    
    • route record / record-known-failing → proceed.
    • route driver-gap → port the step first (Step 2), then proceed.
    • route frozen → STOP, return status: frozen (nothing to record).
  2. Prerequisites met. of record prereq-check --agent <agent> lists the human actions a recording needs (auth mode, env vars, a mock, a local model server). If one is unmet and you can't satisfy it, STOP and return status: prereq_blocked naming the exact blocker — never ask the dispatcher.

  3. Clean replaydata/ tree. The recording precheck refuses a dirty replaydata/ (re-records must be deliberate commits). If the assessment isn't committed yet, that's the dispatcher's ordering bug — return status: infra_fail with that note.

  4. A recording daemon is up — attached OR coexisting. Which one is a decision you make per run, not a default:

    • Attach (--attach, against the user's running irrlichd --record) when you are re-recording an adapter the running daemon already ships, and that daemon is genuinely in --record mode. The dashboard stays connected and the session shows up live.

    • Coexist — an isolated daemon built from this branch, on its own IRRLICHT_HOME and port — when either half of that is false:

      IRRLICHT_ONBOARD_HOME=/tmp/irr-onb \
      IRRLICHT_ONBOARD_BIND_ADDR=127.0.0.1:7838 \
      IRRLICHT_PERMISSION_MODE=grant-all \
        of record run --agent <agent> --scenario <scenario>
      

Read the full file on GitHub · 502 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. 1mo ago Changed · +45 lines scan B → D be3a5bd95fc7
  2. 1mo ago Changed · +7 lines 7292c2f1496a
  3. 1mo ago First seen · 450 lines · 106 tokens per session scan B 8df67121b961

Subscribe to this mod's changes

ir:onboarding-factory/record is a skill published in the GitHub repository ingo-eichhorst/Irrlicht (101 stars, last pushed today), licensed MIT. It adds 106 tokens to every session and 6,588 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it D with 2 findings (reads agent configuration directories, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

eval-dataset-design

A guide to creating reliable test sets for AI agents. Each evaluation task includes the request, the simulated user's behaviour, a reset starting state, and a success check that can be independently verified.

bojieli/ai-agent-book · 138 tokens

verify-omnigent

Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…

omnigent-ai/omnigent · 100 tokens

test-master

Generates test files, creates mocking strategies, analyzes code coverage, designs test architectures, and produces test plans and defect reports across functional, performance, and security testing disciplines. Use when writing unit tests, integration tests, or E2E tests; creating test strategies or automation…

Jeffallan/claude-skills · 104 tokens

dogfood

Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.

callstack/agent-device · 55 tokens

cua-driver

Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.

davidondrej/skills · 68 tokens

critique

Adversarial design critique of the current work. Render it, look at it, and argue for rejection. Run after the gates are green, never instead of them.

plugin87/ux-ui-agent-skills · 37 tokens