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-ideategit 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-ideate)<a href="https://agentmods.dev/skills/ruc-nlpir/arbor/arbor-agent-ideate"><img src="https://agentmods.dev/badge/skills/ruc-nlpir/arbor/arbor-agent-ideate/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/ruc-nlpir/arbor/arbor-agent-ideate"><img src="https://agentmods.dev/badge/skills/ruc-nlpir/arbor/arbor-agent-ideate.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00066 | $0.01213 |
| Opus 5 | $0.00033 | $0.00607 |
| Sonnet 5 | $0.00013 | $0.00243 |
| Haiku 4.5 | $0.00007 | $0.00121 |
Grade A, and why
arbor-agent-ideate 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 9d 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Arbor Ideation Gate
Load this only during IDEATE. The coordinator must first call
TreeView(format="constraints").
After loading, write a brief visible progress note if useful:
LOAD_RECEIPT: # SKILL: Idea Drafting
Hard Sequence
- Read constraints: root insight, pruned lessons, validated findings, tree shape, and sibling nodes.
- Run the Probe Block below before listing candidates.
- Generate candidates using all four idea moves.
- Apply depth-aware abstraction.
- Write a five-field scratch declaration for each survivor.
- Run the pre-submission self-check.
- Commit each survivor with
TreeAddNodeusing exactly the four labelled lines specified below.
If any step is skipped, restart IDEATE.
Probe Block
Answer all four questions with concrete evidence: failure case IDs, log lines, metrics, source references, or experiment reports.
PROBE BLOCK
Q1 First principles : <bottleneck CLASS> - evidence: <case ids / log refs>
Q2 Hidden assumption: <assumption> - if dropped: <what opens up>
Q3 Elephant : <ugly problem the trunk currently ignores>
Q4 Hamming : <yes/no, plus one sentence justification>
Q1 must name a failure class, such as wrong retrieval, wrong reasoning over correct evidence, wrong stopping condition, wrong representation, wrong objective, wrong action space, or wrong credit assignment. If you cannot cite at least two concrete pieces of evidence, go back to OBSERVE.
Mindset
- Think like a principal investigator, not an engineer filing a small PR.
- Ask HOW, not HOW MUCH: change an algorithm, representation, objective, control flow, data structure, or reasoning strategy.
- Aim at a class-level bottleneck, not a single example.
- Mechanism is a noun. "Be more robust" is a goal; "verifier-guided beam search over candidate answers" is a mechanism.
Idea Generation Moves
Use all four moves before selecting candidates:
- Assumption Inversion: take the Q2 assumption and design a mechanism that works when it is false.
- Backward From Success: imagine the benchmark solved; identify missing pipeline stages, state, or signals.
- Analogical Transfer: borrow mechanisms from search, CSP/SAT, debate, program synthesis, control theory, or scientific method.
- Failure-Case Reverse Engineering: pick 2-3 failures and ask what minimal capability would have caught them; cluster the answers.
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.
- 9d ago First seen · 130 lines · 66 tokens per session scan A 1b55bc6f7fc8
arbor-agent-ideate is a skill published in the GitHub repository RUC-NLPIR/Arbor (1,059 stars, last pushed yesterday), licensed Apache-2.0. It adds 66 tokens to every session and 1,213 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.