تشغيل أي مهارة في Manus بنقرة واحدة

spec-interview

النجوم٢

التفرعات٠

آخر تحديث١٢ يوليو ٢٠٢٦ في ٠٥:٠٦

Specification-first workflow for new projects, existing-repo onboarding, unresolved draft specs, and broad or architecture-shaping features. Detect project state before asking questions and resolve only choices that affect implementation. Clear bounded changes do not require an interview.

التثبيت

التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.

تشغيل في Manus

المصدر

eightmm

eightmm/oh-my-setting

فتح مستودع GitHub عرض مستودعات المنشئ

تنزيل

تشغيل في Manus

المهن ذات الصلةSOC

استنادا إلى تصنيف SOC المهني

مطوّرو البرمجياتمهن الحاسوب والرياضيات·SOC 15-1252

مستكشف الملفات

4 ملفات

SKILL.md

readonly

المزيد من هذا المستودع

نفس المستودع

agent-harness

eightmm/oh-my-setting

Shared multi-agent harness state and coordination for Codex, Claude Code, and Antigravity. Use for repo resume/state, shared memory or active-task handoff, plan claims/recovery, fail-ledger checks, stale `.oms` cleanup, role/executor setup, local provider calls or write delegation, patch admission/landing, artifact provenance, and prior-session handoff.

2026-07-122

chem-bio-ml

eightmm/oh-my-setting

Discovery-focused chem-bio ML guardrails for small molecules, molecular 3D and physics, proteins and variants, biomolecular interactions, peptides and antibodies, reactions and synthesis, generative design, cellular omics and phenotypic profiling, and biomedical networks. Use before writing, reviewing, evaluating, or training on RDKit/SMILES/SELFIES, QSAR/ADMET, quantum or force data, MMseqs2/MSA/protein language models, PDB/AlphaFold/DMS, docking/DTI/PPI, peptide-MHC or antibody data, reaction/retrosynthesis/yield data, molecular or protein generation, RNA/oligonucleotide/CRISPR, single-cell perturbation/Cell Painting, or biomedical knowledge graphs. Enforce domain splits, leakage checks, label semantics, baselines, metrics, calibration, applicability domain, and provenance.

2026-07-122

ml-training

eightmm/oh-my-setting

PyTorch training correctness for optimizers and schedules, distributed/DDP, variable-size masked losses, checkpoint save/load, CUDA runtime settings, and equivariant model checks. Use when writing or reviewing training loops, optimizer/scheduler setup, multi-GPU code, loss reduction, or checkpoints.

2026-07-122

peer-review

eightmm/oh-my-setting

Multi-agent code review. Use when the user explicitly requests independent verification, cross-agent/council review, or a release go/no-go or requested ML pre-training gate needs several independent reviewers. High-risk changes alone do not require this skill.

2026-07-122

research-method

eightmm/oh-my-setting

Hypothesis-driven loop for ML and scientific experiments. Use to turn a question into a falsifiable hypothesis, design the smallest disconfirming run, pre-register prediction/baseline/metric, interpret results, coordinate experiment claims, reproduce or compare runs, or trace an output to its run.

2026-07-122

slurm-hpc

eightmm/oh-my-setting

Slurm/HPC workflow helper. Use when working on clusters, partitions, nodes, sbatch/srun jobs, GPU/CPU allocations, queues, logs, checkpoints, or any resource-heavy command — e.g. "submit the job", "check the queue", "잡 돌려줘", "큐 확인", "GPU 몇 장", "학습 돌려줘" on a cluster; also digesting a long job log (job-digest, "로그 요약") and reconciling finished Slurm jobs into shared state (run-reconcile, sacct). Reads local cluster reference when generated.

2026-07-122

name	spec-interview
description	Specification-first workflow for new projects, existing-repo onboarding, unresolved draft specs, and broad or architecture-shaping features. Detect project state before asking questions and resolve only choices that affect implementation. Clear bounded changes do not require an interview.

Specification Interview

Do not code before the task-relevant contract is clear. Do not force an interview for a bounded change with a clear local contract. Bounded changes with a clear local contract do not require an interview.

State Router

State	Route
New project	Interview intent/scope/execution -> draft `PROJECT.md` -> confirm -> bootstrap
Existing repo, not onboarded	Inspect code/config/git -> draft from evidence -> resolve material gaps -> confirm -> apply template -> doctor
Ongoing draft	Read open decisions; ask only when they affect this request
Ongoing confirmed	Read current task/verification; proceed unless spec and reality drifted

Use managed block markers, PROJECT.md state, and the source tree as evidence; do not ask the user which state applies.

Flow

Inspect before asking.
Clarify goal, users, and non-goals, then interface, data, paths, and constraints, then success criteria/verification/resources. Skip questions answered by evidence.
Write or update PROJECT.md for project work; use a compact task spec for non-project work.
List assumptions and unresolved material decisions.
Proceed after confirmation. A new-project bootstrap requires confirmed PROJECT.md; a bounded existing change may proceed from its local contract.

Must resolve uncertainty about public API/CLI, persistence/schema, auth/privacy, destructive or expensive operations, Slurm resources, dependencies, and acceptance criteria when it affects the requested work.

Route to One Reference

Native question UI, fallback format, and question discipline: question-ui.md
New-project template selection and safe bootstrap: project-bootstrap.md
PROJECT.md and compact non-project spec shapes: spec-templates.md

Report detected state, evidence used, decisions captured, remaining blockers, files created/changed, and verification. Never tell the user to run bootstrap commands manually when the current agent can run them.