What is ir:onboarding-factory/assess?

A workflow for judging one coding-agent scenario across three areas: what the agent can do, what the recording system captures, and what the execution driver can perform. It then writes the scenario's metadata and machine-checkable expectations.

Use it to assess one agent-and-scenario combination, write its execution recipe, define expected recording phases and observations, and choose the resulting evaluation route.

It makes evaluation decisions traceable by requiring evidence, confidence, reasoning, caveats, and sources for each area. It also produces checks that recordings can verify automatically.

How ingo-eichhorst/Irrlicht configures Claude Code

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/assess/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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/ingo-eichhorst/irrlicht/assess.svg)](https://agentmods.dev/skills/ingo-eichhorst/irrlicht/assess)

Your own site

<a href="https://agentmods.dev/skills/ingo-eichhorst/irrlicht/assess"><img src="https://agentmods.dev/badge/skills/ingo-eichhorst/irrlicht/assess.svg" alt="Measured on agentmods" height="20"></a>

Per session 93 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 4,956 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

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.

Model	Per session	Once invoked
Fable 5.1	$0.00093	$0.04956
Opus 5	$0.00046	$0.02478
Sonnet 5	$0.00019	$0.00991
Haiku 4.5	$0.00009	$0.00496

Measured 7d ago against content hash bdcf75c4ffba, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

ir:onboarding-factory/assess 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 7d 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/skills/ir:onboarding-factory/assess/SKILL.md · 358 lines

How it starts

The opening of the file, as written. The whole thing — 358 lines — stays where its author put it; the contents beside it link to each section on GitHub.

assess

You run as a focused subagent with no parent context. Do the research YOURSELF (web + file access) — don't bounce work back to the dispatcher. This verb spends NO API tokens on agent CLIs and runs NO recording. When done, return only the "Return contract" block.

What this produces

For one cell it writes two artifacts, both through the factory (never by hand):

The cell — of cell write writes replaydata/agents/<agent>/scenarios/<id>_<scenario>/metadata.json: the three-pillar verdict + confidence + a note, the full reasoning + caveats + sources, and the per-agent recipe (the driver step sequence record will run).
The spec — of cell spec writes expected.jsonl in the same folder: the machine-checkable phases AND the observation assertions (model / cost / tokens / agent) that record's verify step checks against the recording.

The route the dispatcher reads off of status is DERIVED from the three pillars — see the routing table in ../return-contract.md.

The three pillars (judge each, cite each)

Read the pillar definitions in ../return-contract.md. Three rules govern every verdict:

Honest verdicts, anchored to evidence. agent=yes only when the docs/code state the behavior explicitly; no when something fundamental blocks it; unknown over a guess. Name the RIGHT owner on the daemon pillar: bug (product — file an issue) vs incapable (architecture) vs the driver pillar's gap:<primitive> (tooling) each route differently. Don't park ambiguity in bug the way a catch-all partial once was a dumping ground.
Caveats over downgrades. If the canonical spec is met but a narrow detail is gappy, keep daemon=full and put the gap in caveats. A caveat is NOT a bug: reserve bug for a spec-required observable the daemon mis-handles.
Cite primary sources. Agent docs, official changelog, agent source, irrlicht adapter source, and — for bug/incapable — the recording's events.jsonl. Tutorials and blogs don't count. Even an unknown verdict cites what you searched.

Read the full file on GitHub · 358 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.

7d ago First seen · 358 lines · 93 tokens per session scan A bdcf75c4ffba

Subscribe to this mod's changes

ir:onboarding-factory/assess is a skill published in the GitHub repository ingo-eichhorst/Irrlicht (97 stars, last pushed today), licensed MIT. It adds 93 tokens to every session and 4,956 once invoked, about $0.0005 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.

Building agents

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens

chronicle

Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…

microsoft/vscode · 72 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

Irrlicht: Skill for Claude Code