Harbor is a framework for evaluating and improving AI agents and language models across reusable benchmarks and execution environments. Developers and researchers use it to run agent evaluations, create benchmarks, perform parallel experiments, and generate rollouts for reinforcement-learning optimization, with the catalogue offering skills, instructions, and an MCP integration for the framework.
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 harbor-framework/harbor --skill publishgit clone --depth 1 https://github.com/harbor-framework/harborWrote 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/harbor-framework/harbor/publish)<a href="https://agentmods.dev/skills/harbor-framework/harbor/publish"><img src="https://agentmods.dev/badge/skills/harbor-framework/harbor/publish/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/harbor-framework/harbor/publish"><img src="https://agentmods.dev/badge/skills/harbor-framework/harbor/publish.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 21 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 8 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 12 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 16 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 20 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 26 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 31 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 32 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 40 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 53 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 54 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 55 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 56 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 57 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 58 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 66 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 67 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 73 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 76 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 87 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 92 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
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 | $0.00035 | $0.00859 |
| Opus 5.5 | $0.00014 | $0.00344 |
| Sonnet 5.5 | $0.00007 | $0.00172 |
| Haiku 4.5 | $0.00003 | $0.00086 |
Grade A, and why
publish 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 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.
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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Help the user publish a Harbor task or dataset to the registry. Walk them through each step, checking prerequisites and confirming before running publish commands.
All commands use uvx harbor so the user does not need to install the CLI globally.
Prerequisites
-
Auth: Ensure the user is logged in by running
uvx harbor auth status. If not logged in, prompt them to runuvx harbor auth login. -
Task identifiers: All tasks must have a
[task]section in theirtask.tomlwith a name like<org>/<name>. If missing, run:uvx harbor task update "<path/to/task>" --org "<org>"To update all tasks in a directory:
uvx harbor task update "<path/to/tasks>" --org "<org>" --scan
Publishing a task
uvx harbor publish "<path/to/task>"
Publish multiple tasks or all tasks in a directory:
uvx harbor publish "<path/to/task-a>" "<path/to/task-b>"
uvx harbor publish "<path/to/tasks>"
Publishing a dataset
1. Initialize the dataset manifest (if no dataset.toml exists)
uvx harbor dataset init "<org>/<dataset>" \
--description "<description>" \
--author "Jane Doe <[email protected]>"
Add --with-metric if the user needs a custom metric script. If omitted, the dataset uses the default metric (average reward across tasks, with missing values treated as 0).
Auto-add behavior: If dataset init is run in a directory that already contains tasks, those tasks are automatically added to the manifest. After init, ask the user if they want to add additional tasks — either from the Harbor registry or from other local folders.
2. Add additional tasks and files (optional)
cd "<path/to/dataset>"
uvx harbor dataset add "<path/to/task-a>" "<path/to/task-b>" # local tasks from elsewhere
uvx harbor dataset add "<path/to/folder>" --scan # all tasks in another folder
uvx harbor dataset add org/task-name # published tasks from registry
uvx harbor dataset add org/dataset-name # all tasks from a published dataset
uvx harbor dataset add metric.py # metric file (same dir as dataset.toml)
uvx harbor dataset remove "<org>/<task-name>" # remove a task
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.
- 1mo ago Changed 25d7c2ec4a99
- 2mo ago First seen · 97 lines · 35 tokens per session scan A 7a2a581a29ea
publish is a skill published in the GitHub repository harbor-framework/harbor (6,630 stars, last pushed today), licensed Apache-2.0. It adds 35 tokens to every session and 859 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
research-report
Research report on companies, competitors, markets or products: research a company, compare competitors, market landscape, funding, pricing, launches. Also to save a brief, summary or snapshot built from search results.
da-anomaly-analysis
Metric anomaly investigation: DAU, GMV, conversion, retention, revenue, KPI or dashboard metric drop, dropped, fell, spike or spiked. Metric attribution and funnel breakdown.
meeting-prep
Meeting prep: brief me before a meeting, call, standup, 1:1 or pitch — who I'm meeting, background, talking points; prep me, ready.
schedule-meeting
Schedule, book or set up a meeting, call, coffee, lunch, dinner, event or appointment on my calendar.
building-agents
Use when building or restructuring an LLM agent — provider adapter, tool calling, structured output, RAG, agent loop, eval gate, cost routing, tracing, MCP server — model-agnostic across OpenAI/Anthropic/Gemini/OSS so a model swap is a config change. NOT vector-store SQL alone (that is postgresdb) or service…
agent-eval
Use when measuring whether an LLM or agent system actually got better and gating merges on it: golden sets, fixing an inflated LLM-as-judge, scoring RAG (faithfulness, contextual recall) or agent trajectories (tool correctness, completion), or picking an eval framework. NOT building the agent loop, tools or RAG…