Map task scope, datasets, metrics, and baselines for ambiguous quests.
npx skills add https://github.com/Rycen7822/DeepScientist-hermes --skill scout-rycen7822
Please help me install this Agent Skill. Skill: scout Source: https://github.com/Rycen7822/DeepScientist-hermes/tree/main/resources/skills/scout Command: npx skills add https://github.com/Rycen7822/DeepScientist-hermes --skill scout-rycen7822
Scout helps identify task framing and discovery needs when a quest is ambiguous, by surfacing framing questions, relevant literature, and a durable evaluation contract.
Begin by clarifying the quest objective, then assemble a framing report and a baseline shortlist to proceed.
High-intent search queries and answers about installing and using this skill.
Framing an ambiguous research quest requires mapping the task scope, datasets, metrics, and baselines. You clarify the objective, assemble a framing report, and produce a baseline shortlist to guide next steps.
Literature scouting maps task-relevant papers, benchmarks, and official repos to establish a durable evaluation contract. It enforces disciplined memory and known references to define metrics and baselines for your quest.
Creating a baseline shortlist involves performing literature reconnaissance to find known references and benchmarks. You map these to your task scope to establish a clear evaluation contract for next steps.
You need to map task scope and datasets when your quest objective is unclear and the repo's literature is not yet defined. This framing identifies task discovery needs and surfaces relevant evaluation contracts.
The best way to structure durable artifacts is to produce a brief, a plan, and a literature scouting report. These artifacts capture the framing questions and baseline shortlist to guide subsequent anchors.
Yes, you can use scouting to evaluate task discovery needs without prior benchmarks. It surfaces framing questions and performs literature reconnaissance to map task-relevant papers and establish a baseline shortlist.
qa-amanDraft comprehensive BRDs from user input with objectives, scope, constraints, and risks.
tadod12Automates SAE training/evaluation to decompose model activations into interpretable features using SAELens metrics.
stevolandCreate and manage MADR-compliant Architecture Decision Records across projects.
BoomberAspExplore literature and baselines to frame early-stage research tasks.
