Arbor is an autonomous research agent that proposes hypotheses, edits code, runs experiments, and retains improvements that succeed on held-out data in a growing hypothesis tree. Researchers use it to investigate problems and iteratively optimize solutions with real experiments. The catalogue skills and plugin expose Arbor's research-agent workflow to coding agents.
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 RUC-NLPIR/Arbor --skill arbor-agent-toolsgit clone --depth 1 https://github.com/RUC-NLPIR/ArborWrote 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/ruc-nlpir/arbor/arbor-agent-tools)<a href="https://agentmods.dev/skills/ruc-nlpir/arbor/arbor-agent-tools"><img src="https://agentmods.dev/badge/skills/ruc-nlpir/arbor/arbor-agent-tools.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00069 | $0.00965 |
| Opus 5 | $0.00034 | $0.00483 |
| Sonnet 5 | $0.00014 | $0.00193 |
| Haiku 4.5 | $0.00007 | $0.00097 |
Grade A, and why
arbor-agent-tools scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Arbor Agent Tools
Use this skill when the host does not provide native Arbor tools. The bundled script stores state in the same style as open-source Arbor:
<cwd>/.arbor/sessions/<run_name>/.coordinator/idea_tree.json
<cwd>/.arbor/sessions/<run_name>/.coordinator/idea_tree.md
Script
scripts/arbor_state.py is stdlib-only.
Common commands:
TOOLS="<skill-dir>/arbor-agent-tools/scripts/arbor_state.py"
python "$TOOLS" init --cwd <project> --run-name <run> --task "<contract>"
python "$TOOLS" view --cwd <project> --run-name <run> --format constraints
python "$TOOLS" meta --cwd <project> --run-name <run> --set baseline_score=42 --set trunk_score=42
python "$TOOLS" meta --cwd <project> --run-name <run> --set "eval_cmd=cd {cwd} && bash eval.sh"
python "$TOOLS" add --cwd <project> --run-name <run> --parent-id ROOT --hypothesis "<four-line hypothesis>"
python "$TOOLS" update --cwd <project> --run-name <run> --node-id 1 --status done --score 45 --insight "..."
python "$TOOLS" worktree --cwd <project> --run-name <run> --node-id 1 --trunk <trunk_branch>
python "$TOOLS" prompt-executor --cwd <project> --run-name <run> --node-id 1 --workdir <worktree>
python "$TOOLS" prompt-executor --cwd <project> --run-name <run> --node-id 1 --smoke
python "$TOOLS" eval --cwd <project> --run-name <run> --split dev --exec-cwd <worktree> --cmd "bash {cwd}/eval.sh" --set-meta trunk
python "$TOOLS" record --cwd <project> --run-name <run> --node-id 1 --score 45 --insight "..." --result "..."
python "$TOOLS" parse-log --log <project>/run.log --metric val_bpb
python "$TOOLS" report --cwd <project> --run-name <run>
python "$TOOLS" check --cwd <project> --run-name <run> --require-report --require-experiment --require-executor-prompt
Read references/tool-mapping.md when deciding which script command maps to a
native Arbor tool.
State Rules
- Keep scores absolute.
- Keep eval commands templated with
{cwd}and{node_id}. - Do not run B_test during executor iteration.
- Use
recordfor executor outcomes so artifacts and tree updates stay in one place. - Use
checkbefore trusting a hand-edited tree. Add artifact flags such as--require-report,--require-experiment,--require-executor-prompt,--require-events,--require-run-stats, or--strict-artifactswhen validating a completed run. - Serialize tree-mutating commands for the same run. Do not parallelize
init,meta,add,update,prune,propagate,eval,record,worktree, ormerge.
What ships with it
3 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 · 75 lines · 69 tokens per session scan A e38bcb844489
arbor-agent-tools is a skill published in the GitHub repository RUC-NLPIR/Arbor (1,058 stars, last pushed 8d ago), licensed Apache-2.0. It adds 69 tokens to every session and 965 once invoked, about $0.0003 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
skill-forge
Autonome Verbesserung nach dem Autoresearch-Paradigma (Karpathy). Zwei Modi: (1) Skill-Modus — optimiert eine SKILL.md durch iterative Mutation und Evaluation. (2) Generic-Modus — optimiert beliebige Dateien gegen jede mechanische Metrik (Testabdeckung, Bundle-Size, Lighthouse-Score, Docker-Image-Größe, etc.). Zwei…
autogpt-agents
Autonomous AI agent platform for building and deploying continuous agents. Use when creating visual workflow agents, deploying persistent autonomous agents, or building complex multi-step AI automation systems.
autocontext-consumer
Use when an agent needs to USE knowledge Autocontext already produced - find which scenarios have knowledge, read the playbook and lessons for one, understand the on-disk file and folder layout, and move knowledge between checkouts. Host-agnostic; requires only the autoctx CLI and the filesystem.
autocontext-creator
Use when an agent needs to CREATE knowledge with Autocontext - run a scenario or plain-language task through the improvement loop, judge or improve a single output, and inspect what the run produced. Host-agnostic; requires only the autoctx CLI.
autocontext
Iterative strategy generation and evaluation system. Use when the user wants to evaluate agent output quality, run improvement loops, queue tasks for background evaluation, check run status, inspect runtime artifacts and session branch lineage, or discover available scenarios. Provides LLM-based judging with…
codex-autoresearch
Triage improvement work and run or resume accepted measured loops in a local project. Architecture, documentation, UX, product study, open research, taste, and one-shot fixes stay direct unless the user explicitly requests repeated measurement with a complete experiment contract.