一键在 Manus 中运行任何 Skill

phase1-exploration

星标6

分支1

更新时间2026年7月7日 07:52

Run Phase 1 (EXPLORATION) of the A2MC calibration workflow as the offline agent — the human-in-the-loop analog of the orchestrator's `_run_exploration()` / `reasoning.analyze_sensitivity_results()`. Extract the Y matrix from completed simulations, run Morris sensitivity analysis, and interpret μ* to decide what parameters matter for each target/PFT. Use when the user says "run the sensitivity analysis", "extract the Y matrix", "which parameters matter", "run Phase 1", "Morris rankings", after the Phase 0 ensemble finishes.

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

jingtao-lbl

jingtao-lbl/A2MC-elm

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

Phase 1: Exploration (offline agent)

The offline analog of _run_exploration(). Online, reasoning.analyze_sensitivity_results() interprets the Morris output automatically; offline, YOU interpret it — attribute each ranked parameter to a FATES mechanism, spot cross-PFT vs PFT-specific importance, and decide what's worth tuning. The extraction + Morris scripts are the same the online agent uses.

Floor, not ceiling. Morris μ* is the online agent's one lens. As the offline agent, reach for more when the round warrants it: check σ/μ* for interaction-heavy parameters, look for edge effects (params pinned near bounds → a Phase 0 redesign signal), compute a second output variable the loop didn't rank, or run a targeted correlation scientific-analysis the fixed pipeline has no step for. Produce the ranking the online agent would, then go past it.

Inputs (from Phase 0): completed TRANS outputs, the X matrix ($A2MC_ENSEMBLE_MATRIX_FILE), the SALib problem ($A2MC_SALIB_PROBLEM_FILE). Deliverable: Morris rankings (μ, μ*, σ per parameter per PFT) as CSV + plots, plus your interpretation of what to tune (captured in the phase log), handed to Phase 2.

Step 0 — completion gate

tools/diagnose_ensemble_status.py --cases 1-N — proceed when >95% of cases have TRANS complete. Failed cases become NaN rows downstream.

Step 1 — extract the Y matrix (reads existing outputs — no new simulations)

phases/phase1_exploration/extract_sensitivity_outputs.py --output-var leaf_biomass --cases 1-N --validation-period 2010 2019 → Morris{Var}_{N}cases.txt (cases × PFTs). Runs on the login node; large ensembles can be memory-heavy — extract in --cases ranges, and use --resume to continue an interrupted run.

Footgun — row-order alignment. The X-matrix row order MUST match the Y-matrix row order (same cases, same order) or SALib Morris is meaningless. If you filter failed cases, filter both sides with completed_cases_<TS>.txt. Audit for NaN rows — >5% NaN, investigate failures first.

Step 2 — run Morris (analysis of the extracted matrices — no new simulations)

phases/phase1_exploration/morris_sensitivity_analysis.py --output-var leaf_biomass --y-matrix <Y> --x-matrix <X> --problem <salib_problem> → per-PFT morris_*.csv (parameter, μ, μ*, σ, rank) + color-coded *.png. analyze_ensemble.py is the higher-level driver that chains extract → Morris.

Step 3 — interpret (you are the reasoning)

Produce the same shape as the online analyze_sensitivity_results output: key_parameters (each attributed to a FATES mechanism — verify against RAG, never from the name), interactions (high σ/μ*), cross_pft_patterns (generic-to-all vs PFT-specific), edge_effects (near-bound → redesign candidates), and a calibration strategy (tune order, PFT-by-PFT vs global).

Step 4 — log and hand off

Log via calibration-log (phase log → PhaseLogger.log_exploration): top parameters per PFT, the mechanism attributions, edge effects, and the tuning recommendation. Hand off to phase2-screening.

Related skills / next phase

Standardized single-round / cross-round figures → summarize-calibration-round, compare-calibration-rounds (they invoke the same Morris pipeline).
A one-off sensitivity question with a figure + ana_log → scientific-analysis.
Next: phase2-screening (rank the ensemble against targets).

Changelog

2026-07-02: Created — offline Phase 1 routine mirroring reasoning.analyze_sensitivity_results(); drives extract_sensitivity_outputs → morris_sensitivity_analysis, hands off to phase2-screening.

同仓库更多 Skills

同仓库

a2mc-init

jingtao-lbl/A2MC-elm

First-run interactive setup for the offline agent — the official "getting started" flow the FIRST time someone uses A2MC in a repo/site. Interview the user (is a2mc_config.sh set up, where is the E3SM/ELM-FATES checkout, FATES on/off, carbon-only vs nutrient-enabled PARTEH, ECA vs RD, site name/location, PFTs, calibration targets), verify the checkout + RAG milestone, then create and populate the use case (site config + validation targets + parameter list) and hand off to phase0-design. Use when the user says "set up A2MC", "first time using A2MC", "help me get started / onboard me to A2MC", "initialize a new site / use case", "configure A2MC for my site", "I want to calibrate a new site". DISTINCT from onboard-session (which resumes an ALREADY-configured setup).

2026-07-076

add-skill

jingtao-lbl/A2MC-elm

Scaffold + register (or retire + de-register) an A2MC skill so the registry steps are never half-done. Use when the user or agent wants to create a new capability — "add a skill", "create/scaffold a skill", "make this reusable as a skill", "distill X into a skill" — OR to remove one — "remove/retire/delete the X skill". Writes (or deletes) the SKILL.md, and registers (or de-registers) it across all THREE human-facing registries (the README 'Current skills' table, docs/a2mc_reference/skills_catalog.md, AND the AGENTS.md 'At a glance' table), runs the mechanical drift check, and stops for human review before commit. Human-gated.

2026-07-076

arm-hpc-monitoring

jingtao-lbl/A2MC-elm

Set up real-time monitoring of an active A2MC HPC ensemble or experiment on Perlmutter at session start (CLAUDE.md Rule

2026-07-076

calibration-log

jingtao-lbl/A2MC-elm

Log interactive calibration work for a site under use_cases/{site}/memory/logs/. Two log types — a PHASE log (a diagnosis, screening, experiment design/result, refinement — recorded the same way the autonomous agent does, via PhaseLogger, so both modes' logs synthesize together) or a free-form SESSION log (exploratory work that does not map to a specific phase). Use when the user wants to record calibration or exploration work — "log this", "log this phase / diagnosis / experiment", "log this calibration session", "write a session log", "record what I explored".

2026-07-076

compare-calibration-rounds

jingtao-lbl/A2MC-elm

Compare A2MC calibration rounds (R1, R2, …) against each other and the validation targets — top-N best-achievable biomass per round, per-target Morris μ* sensitivity, P-pool / cross-regime overlays. Use when the user asks to "compare rounds", "update the multi-round figure", "refresh the R1-RN comparison", "which round is best", "top sensitive parameters per round", or to regenerate any `multiround_*` / cross-round figure after new cases complete. Codifies the multiround bundle pipeline + the footguns (screening contamination, case-name patterns, μ* ranking, param-set mismatch, partial-ensemble caveats).

2026-07-076

inject-knowledge

jingtao-lbl/A2MC-elm

Inject a human-originated domain fact (a discovery, a parameter insight, a mechanism/relationship) into A2MC's curated knowledge so the calibration agent surfaces it — placing it correctly across the up-to-three channels (site discoveries.json, generic parameters.json, curated_relationships.yaml), validating, and rebuilding the graph. Use when the user says "add this finding/parameter knowledge to A2MC's AI", "inject the X discovery into the KB", "make the agent aware of X", "author a curated relationship for X", or wants a manuscript/literature insight to reach the reasoning pipeline. This is the HUMAN-originated counterpart to curate-knowledge (which promotes run-originated proposals); both are Tier-3 curated writes and you (the human-in-the-loop) own the truth call. Distilled from graphrag_curated_yaml_roadmap.md + dev log 20260519d (the clumping_index injection).

2026-07-076

name	phase1-exploration
description	Run Phase 1 (EXPLORATION) of the A2MC calibration workflow as the offline agent — the human-in-the-loop analog of the orchestrator's `_run_exploration()` / `reasoning.analyze_sensitivity_results()`. Extract the Y matrix from completed simulations, run Morris sensitivity analysis, and interpret μ* to decide what parameters matter for each target/PFT. Use when the user says "run the sensitivity analysis", "extract the Y matrix", "which parameters matter", "run Phase 1", "Morris rankings", after the Phase 0 ensemble finishes.
modes	{"requires_fates":false,"nutrient_pathway":"any","scope":["calibration"],"summary":"Offline analog of Phase 1 (extract Y matrix, Morris sensitivity). Applies in every calibration mode."}

Phase 1: Exploration (offline agent)

Floor, not ceiling. Morris μ* is the online agent's one lens. As the offline agent, reach for more when the round warrants it: check σ/μ* for interaction-heavy parameters, look for edge effects (params pinned near bounds → a Phase 0 redesign signal), compute a second output variable the loop didn't rank, or run a targeted correlation scientific-analysis the fixed pipeline has no step for. Produce the ranking the online agent would, then go past it.

Step 0 — completion gate

tools/diagnose_ensemble_status.py --cases 1-N — proceed when >95% of cases have TRANS complete. Failed cases become NaN rows downstream.

Step 1 — extract the Y matrix (reads existing outputs — no new simulations)

Footgun — row-order alignment. The X-matrix row order MUST match the Y-matrix row order (same cases, same order) or SALib Morris is meaningless. If you filter failed cases, filter both sides with completed_cases_<TS>.txt. Audit for NaN rows — >5% NaN, investigate failures first.

Step 2 — run Morris (analysis of the extracted matrices — no new simulations)

Step 3 — interpret (you are the reasoning)

Step 4 — log and hand off

Related skills / next phase

Standardized single-round / cross-round figures → summarize-calibration-round, compare-calibration-rounds (they invoke the same Morris pipeline).
A one-off sensitivity question with a figure + ana_log → scientific-analysis.
Next: phase2-screening (rank the ensemble against targets).

Changelog

2026-07-02: Created — offline Phase 1 routine mirroring reasoning.analyze_sensitivity_results(); drives extract_sensitivity_outputs → morris_sensitivity_analysis, hands off to phase2-screening.