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 skills add jame581/LogseqBrain --skill evalgit clone --depth 1 https://github.com/jame581/LogseqBrainWrote 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/jame581/logseqbrain/eval)<a href="https://agentmods.dev/skills/jame581/logseqbrain/eval"><img src="https://agentmods.dev/badge/skills/jame581/logseqbrain/eval/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.
<a href="https://agentmods.dev/skills/jame581/logseqbrain/eval"><img src="https://agentmods.dev/badge/skills/jame581/logseqbrain/eval.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00050 | $0.00457 |
| Opus 5.5 | $0.00020 | $0.00183 |
| Sonnet 5.5 | $0.00010 | $0.00091 |
| Haiku 4.5 | $0.00005 | $0.00046 |
Grade A, and why
eval 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 today.
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.
What it actually says
Run the eval suite
evals/README.md is the source of truth. Read it before running anything the first time in a session.
Hard rules
- Run it only through WSL2 as the
logseq-evaluser:wsl -d FedoraLinux-44 -u logseq-eval sh /mnt/d/AI/logseq-brain/tools/eval/run.sh …. Never run it on native Windows, and never as the Docker-linked user. - It evaluates committed files only. If
git statusshows uncommitted changes toplugin/, say that they won't be evaluated. - Run it through PowerShell, or through Bash with single quotes. wsl.exe heredocs halve backslashes.
Modes
| Ask | Command args | Cost |
|---|---|---|
| default / "check" | --check |
free: case schema and tool grants |
| harness self-test | run sh tools/eval/test.sh locally |
free |
| one case | --case <name> |
paid, a fraction of a full run |
| full | (no args) | paid, ≈ $4–7 of plan usage |
Always run --check first. Before any paid mode, state the estimated cost and wait for a yes.
Reading the result
- Exit 0 means every case passed and the canary is clean. A usage-limit error is not a failure; rerun after the limit resets.
- A failed case may be rerun once with
--case <name>. A pass on the rerun counts as a flake, and a second failure is a real failure. - Report per-case pass/fail, both canary lines, and the
load-digestandsave-basictool-call figures. Reports land inevals/results/, which is gitignored.
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.
- today Changed 49811dac969c
- 25d ago First seen · 33 lines · 50 tokens per session scan A 3251f974715c
eval is a skill published in the GitHub repository jame581/LogseqBrain (18 stars, last pushed today), licensed MIT. It adds 50 tokens to every session and 457 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-28.
Other skills, from other repositories
bad-case-to-dpo
A guide to turning failed production agent runs into training and evaluation data. It explains how to find the first mistake, create preference examples for DPO (Direct Preference Optimization), and check that a fix does not damage other tasks.
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.
adk-go-self-review
Review an ADK Go change the way a maintainer will — a fresh-context pass over the whole diff, five lenses (correctness and tests, scope, simplicity, style, adk-python parity), and the mutation check that proves your tests pin the change. Use before opening a PR, before any later push that changes code, and when asked…
release-preflight
Run the release hard-precondition gate before any tag/OTA — apc release preflight runs the real tier gate (run.ts), version alignment, and prisma regen check, all read-only. green(0) releasable · staged(3) fixable preconditions · blocked(1) tier red/envblocked. No push, no write.
webapp-qa
Perform scoped exploratory QA of a web application, reproduce defects, collect screenshots and console evidence, and deliver a prioritized report.
branch_health
Quick health check -- test counts and file stats for AIPass branches.