AI Research Explore (RigorPilot)
Governed exploration of deep-learning research candidates — two-loop rhythm, frozen task/dataset/benchmark, idea gating and fair-comparison ranking, auditable evidence in explore_outputs/
- What
- Governed exploration of deep-learning research candidates — two-loop rhythm, frozen task/dataset/benchmark, idea gating and fair-comparison ranking, auditable evidence in explore_outputs/
- Cost
- Free
- Needs
- a deep-learning research setup — Python 3.11+ and git for the suite's bundled orchestration scripts; a `current_research` anchor (branch, commit, checkpoint or run record) and explicit exploration authorization are needed at use time. Install the companion `ai-research-reproduction` skill too — shared references/ and scripts/ live there.
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor — the RigorPilot "Rigor Explore" orchestration skill for meaningful and potentially novel deep-learning research candidates, with scientific rigor, comparability and reproducibility baked in. It only applies when the researcher has chosen the task family, dataset, benchmark, evaluation method and SOTA references, and explicitly authorizes candidate-only exploration on top of a durable `current_research` anchor. The core is a two-loop research rhythm: an outer loop (understand the repo, freeze task/dataset/evaluation/budget, preserve researcher ideas, map sources, gate ideas, decide whether the next experiment is worth running) and an inner loop (make one bounded candidate change or run, smoke-check it, collect evidence, rank it against the anchor, stop or return to the outer loop). Campaign mode freezes the task, dataset, benchmark, evaluation source, SOTA reference and compute budget before any candidate work; ideas are preserved from the researcher and optionally extended with a small bounded set of single-variable seeds, ranked with explicit gates and score breakdowns; execution prefers one clear candidate at a time, delegating to `explore-code`/`explore-run`, and writes `explore_outputs/` with `SCIENTIFIC_CHANGELOG.md` and `COMPARABILITY_REPORT.md`. Critically, novelty and significance remain hypotheses until literature contrast, ablation evidence and fair comparison — the skill never presents exploratory gains as trusted reproduction success. By @lllllllama, listed here with credit to its creator. Honest caveats: part of the RigorPilot suite — it references shared references/ and scripts/ that live in the companion `ai-research-reproduction` skill, so install the suite together; candidate-only work requires explicit researcher authorization, it's not an autonomous discovery loop; it promises no novelty proof, no global benchmark completeness, no SOTA guarantees. Skill Harbor never reviews the code, review it yourself before use.
Version:
Install
Prerequisites: a deep-learning research setup — Python 3.11+ and git for the suite's bundled orchestration scripts; a `current_research` anchor (branch, commit, checkpoint or run record) and explicit exploration authorization are needed at use time. Install the companion `ai-research-reproduction` skill too — shared references/ and scripts/ live there. Install "AI Research Explore (RigorPilot)" for me. The RigorPilot Rigor Explore orchestration skill: governed exploration of deep-learning research candidates with a two-loop rhythm, frozen task/dataset/benchmark, idea gating, fair-comparison ranking, and auditable evidence written to explore_outputs/. Repository: https://github.com/lllllllama/rigorpilot-skills/blob/main/skills/ai-research-explore/SKILL.md 1. Fetch the SKILL.md file (and any helper files) from the repository path into a temporary folder and summarize what it does in one or two sentences. 2. Safety check: review the SKILL.md and scripts for anything suspicious (unexpected network calls, shell commands, credential harvesting). This repo should contain zero secrets in code, credentials only via the secure vault. Verify that holds here; STOP on any red flag and tell me. 3. Install it as a skill: copy SKILL.md and its helper files into the agent's skills directory, in a folder named "rigorpilot-ai-research-explore". 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself. GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.