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 mifunedev/openharness --skill health-checkgit clone --depth 1 https://github.com/mifunedev/openharnessWrote 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/mifunedev/openharness/health-check)<a href="https://agentmods.dev/skills/mifunedev/openharness/health-check"><img src="https://agentmods.dev/badge/skills/mifunedev/openharness/health-check/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/mifunedev/openharness/health-check"><img src="https://agentmods.dev/badge/skills/mifunedev/openharness/health-check.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 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 Privilege Escalation · line 43 Potential security issue detected. Manual review is recommended.Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
- medium Excessive Agency · line 18 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 262 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00153 | $0.05028 |
| Opus 5 | $0.00077 | $0.02514 |
| Sonnet 5 | $0.00031 | $0.01006 |
| Haiku 4.5 | $0.00015 | $0.00503 |
Grade A, and why
health-check 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 10d 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 — 267 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Health Check
Report-first resource triage. Answers two questions: can we start <target> right now, and what is the safest way to recoup headroom if not. Destructive reclaim is never automatic — only the regenerable build cache is pruned without asking.
This skill spans two scopes, and step 0 decides which one you are in. Docker inventory and every reclaim lever live on the machine running the daemon; memory, swap, disk and CPU can be read on either side but mean different things depending on where you read them. The sandbox has had no host Docker socket since #756 removed it as a host-root escape path, so a sandboxed invocation reports container-scope metrics and hands the Docker half to the orchestrator. Never present container figures as host figures — that is the failure mode this structure exists to prevent, and it is worse than reporting nothing, because a green container disk row reads like permission to start a multi-GB image build.
target is free text naming what you intend to start (a compose path, make dev, a service name). It sizes the verdict — a full docker compose build and a make dev against already-running services have very different footprints, so always pin down which before judging "sufficient."
Performance rules
Default to a fast path: one scope classification, one host snapshot, one Docker summary, one running-container list, and one exited-container list. Do not run verbose or per-container probes until the summary shows a likely binding constraint or the user asks for reclaim candidates.
- Gate expensive commands:
docker system df -v,docker stats,docker exec <container> ps ...,du, and nested-Docker inspection are second-pass diagnostics, not baseline checks. - Run expensive diagnostics once and reuse the captured output. Never call
docker system df -vinside a loop over volumes/images/containers. - Prefer scoped reads over broad scans: explicit cache paths (
~/.npm, project.pnpm-store) are acceptable;du -sh /home/*or filesystem-widefindis not. - Keep reruns delta-oriented: if a prior health-check in the same thread already found the shape of the problem, recheck only the changed metrics unless the target changed.
- Never retry a Docker call that failed for want of a daemon. Step 0 settles reachability once. Nine copies of the same connection error is not diagnosis.
What ships with it
1 file 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.
- 10d ago First seen · 267 lines · 153 tokens per session scan A 2b6101010d6f
health-check is a skill published in the GitHub repository mifunedev/openharness (37 stars, last pushed 2d ago), licensed Apache-2.0. It adds 153 tokens to every session and 5,028 once invoked, about $0.0008 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
tao-setup-nvidia-gpu-host
Host setup for TAO GPU backends. Checks and, after user approval, installs minimum-compatible NVIDIA driver, CUDA Toolkit, and NVIDIA Container Toolkit versions for Docker/local-Docker and Kubernetes GPU worker hosts. TAO-wide defaults can be overridden by the selected model's runtime profile. The --check-only path…
data-visualization
Use for creating publication-quality charts and multi-panel analysis summaries. Triggers when tasks involve visualizing data, plotting results, creating charts, or producing visual reports from analysis output.
cuml-machine-learning
Use for GPU-accelerated machine learning on tabular data using NVIDIA cuML. Triggers when tasks involve classification, regression, clustering, dimensionality reduction, or model training on datasets.
remember
Review the current conversation and capture valuable knowledge — best practices, coding conventions, architecture decisions, workflows, and user feedback — into persistent memory (AGENTS.md) or reusable skills. Use when the user says: (1) remember this, (2) save what we learned, (3) update memory, (4) capture…
schema-exploration
Lists tables, describes columns and data types, identifies foreign key relationships, and maps entity relationships in a database. Use when the user asks about database schema, table structure, column types, what tables exist, ERD, foreign keys, or how entities relate.
deepagents-thread-inspector
Inspect and explain conversations in the local Deep Agents Code SQLite session store. Use as a fallback when LangSmith trace tooling is unavailable, for offline or untraced sessions, or when asked to identify or summarize a local dcode thread, inspect checkpoint metadata, list recent local threads, or parse…