Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/nearform/lastlightnpx agentmods add skills/nearform/lastlight/lastlight-evalsWrote 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/skills/nearform/lastlight/lastlight-evals)<a href="https://agentmods.dev/skills/nearform/lastlight/lastlight-evals"><img src="https://agentmods.dev/badge/skills/nearform/lastlight/lastlight-evals.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to medium
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 →
- medium MCP Rug Pull · line 165 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 166 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 214 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
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 | $0.00185 | $0.04965 |
| Opus 5 | $0.00093 | $0.02482 |
| Sonnet 5 | $0.00037 | $0.00993 |
| Haiku 4.5 | $0.00018 | $0.00496 |
Grade A, and why
lastlight-evals 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 8d 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 — 309 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Set up & run Last Light Evals
lastlight-evals runs Last Light's real production workflows (issue-triage,
build, …) end-to-end against a mocked GitHub, grades the results deterministically
(no LLM-as-judge), and compares models on pass rate, cost, and latency. It's a
thin CLI on top of the lastlight core package (via the lastlight/evals
barrel), so it exercises the same workflows/skills production does. SWE-bench
compatible. Node 24+.
Start here — what do you want to do?
If the user's evals goal is clear, jump to the section. If it's vague ("help me
with the evals", "what can this do?"), ask (AskUserQuestion) which of these
they want, then go:
| Goal | Section |
|---|---|
| Set up an evals workspace (first time) | §2 Scaffold (+ §1 prereqs, §3 providers) |
| Run evals / compare models | §4 Run |
| Look at past runs (no models run) | §4 — lastlight-evals serve |
| Author one case from a GitHub PR or issue | §6 |
| Build a PR-review dataset from my own gold PRs | §6 → "Build a PR-review dataset" |
| Add cases by hand / understand the schema | §5 + references/instance-schema.md |
| Iteratively improve the score toward a target | the lastlight-evals-loop skill |
New to the whole plugin (server / overlay / client, not just evals)? That's the
lastlight-guide skill — this one is evals-only.
1. Check prerequisites
node --version # need >= 24
command -v lastlight-evals >/dev/null && echo "installed" || npm i -g lastlight-evals
lastlight-evals --version # prints the evals version + the bundled lastlight core version
# Optional: for `--sandbox gondolin` (isolate the agent in a QEMU micro-VM so it
# can't read host gold data) you also need QEMU natively: `brew install qemu`
# (macOS) or your distro's `qemu-system` package. The default `--sandbox none`
# needs nothing extra.
2. Scaffold a workspace
init is non-interactive when there's no TTY (piped/agent/CI) — it never
blocks on a prompt. Two layouts:
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 8d ago First seen · 309 lines · 185 tokens per session scan A a50e228035a7
lastlight-evals is a skill published in the GitHub repository nearform/lastlight (22 stars, last pushed 4d ago), licensed MIT. It adds 185 tokens to every session and 4,965 once invoked, about $0.0009 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 skills, from other repositories
local-test
Build, run, and test IronClaw locally using Docker containers and Chrome MCP browser automation.
qa-review
QA review for code changes — test coverage analysis, edge case identification, test plan generation, regression detection, test health tracking over time.
web-ui-test
Test the IronClaw web UI using the Claude for Chrome browser extension.
frontend-engineer
Pro frontend engineering discipline. Enforces build-test-verify workflow for every web project. Never declare done until the site is built, tested, responsive, accessible, and visually verified in a real browser. Use alongside vercel-cli for production-quality deployments.
benchmark-design
Design and calibrate a multi-Case capability Benchmark and establish a traceable Formal Baseline.
agent-optimization
Improve an Agent State through versioned scores and score-linked Traces from a frozen Benchmark.