{"id":"bd7b9604-7f36-4528-93fe-bd5cef22b588","entityType":"agent","slug":"clawhub-ning-kun-amber-md","name":"Amber Molecular Dynamics Simulation","canonicalUrl":"https://www.xpersona.co/agent/clawhub-ning-kun-amber-md","canonicalPath":"/agent/clawhub-ning-kun-amber-md","generatedAt":"2026-10-15T03:21:41.443Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-11T20:54:34.055Z","emptyReason":null},"description":"Official-style Amber24 molecular dynamics workflow guide for proteins. Includes a standard end-to-end Amber MD procedure, command templates, input-file templ... Skill: Amber Molecular Dynamics Simulation Owner: ning-kun Summary: Official-style Amber24 molecular dynamics workflow guide for proteins. Includes a standard end-to-end Amber MD procedure, command templates, input-file templ... Tags: amber:1.0.4, latest:1.0.4, md:1.0.4, molecular-dynamics:1.0.4, simulation:1.0.4 Version history: v1.0.4 | 2026-04-25T10:24:54.717Z | user v1.0.4: upgraded public documentation into an o","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 994 downloads reported by the source. Last updated 10/11/2026.","installCommand":"clawhub skill install s1793t9ymha9akcjnvdf34p37584zr5m:amber-md","sourceUrl":"https://clawhub.ai/ning-kun/amber-md","homepage":"https://clawhub.ai/ning-kun/skills/amber-md","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/ning-kun/amber-md","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/ning-kun/skills/amber-md","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":60,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Official-style Amber24 molecular dynamics workflow guide for proteins. Includes a standard end-to-end Amber MD procedure, command templates, input-file templ..."},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-11T20:54:34.055Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T20:54:34.055Z","emptyReason":null},"stars":null,"forks":null,"downloads":994,"likes":null,"task":null,"library":null,"packageName":null,"latestVersion":"1.0.4","tractionLabel":"994 downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T20:54:33.988Z","emptyReason":null},"lastUpdatedAt":"2026-10-11T20:54:34.055Z","lastCrawledAt":"2026-10-11T20:54:33.988Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-12T20:54:33.988Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.4","createdAt":"2026-04-25T10:24:54.717Z","changelog":"v1.0.4: upgraded public documentation into an official-style manual page with standard command templates and a complete step-by-step manual example.","fileCount":5,"zipByteSize":6477},{"version":"1.0.3","createdAt":"2026-04-25T10:13:55.856Z","changelog":"v1.0.3: converted to documentation-only skill; removed executable scripts to eliminate suspicious scan patterns.","fileCount":4,"zipByteSize":3233},{"version":"1.0.2","createdAt":"2026-04-25T10:09:41.213Z","changelog":"v1.0.2: Eliminated VirusTotal-flagged patterns — extracted inline Python to standalone .py scripts, replaced curl with wget, refactored analysis output.","fileCount":14,"zipByteSize":14190},{"version":"1.0.1","createdAt":"2026-04-25T09:59:57.761Z","changelog":"Full English localization (v1.0.1)","fileCount":11,"zipByteSize":13214},{"version":"1.0.0","createdAt":"2026-04-25T09:45:45.643Z","changelog":"首次发布","fileCount":11,"zipByteSize":13597}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s1793t9ymha9akcjnvdf34p37584zr5m:amber-md","setupComplexity":"low","setupSteps":["Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-ning-kun-amber-md/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-ning-kun-amber-md/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-ning-kun-amber-md/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-ning-kun-amber-md/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-ning-kun-amber-md/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-ning-kun-amber-md/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-15T03:21:41.443Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-ning-kun-amber-md/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-ning-kun-amber-md/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-ning-kun-amber-md/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-ning-kun-amber-md/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"high","updatedAt":"2026-10-11T20:54:34.055Z","emptyReason":null},"readme":"Skill: Amber Molecular Dynamics Simulation\n\nOwner: ning-kun\n\nSummary: Official-style Amber24 molecular dynamics workflow guide for proteins. Includes a standard end-to-end Amber MD procedure, command templates, input-file templ...\n\nTags: amber:1.0.4, latest:1.0.4, md:1.0.4, molecular-dynamics:1.0.4, simulation:1.0.4\n\nVersion history:\n\nv1.0.4 | 2026-04-25T10:24:54.717Z | user\n\nv1.0.4: upgraded public documentation into an official-style manual page with standard command templates and a complete step-by-step manual example.\n\nv1.0.3 | 2026-04-25T10:13:55.856Z | user\n\nv1.0.3: converted to documentation-only skill; removed executable scripts to eliminate suspicious scan patterns.\n\nv1.0.2 | 2026-04-25T10:09:41.213Z | user\n\nv1.0.2: Eliminated VirusTotal-flagged patterns — extracted inline Python to standalone .py scripts, replaced curl with wget, refactored analysis output.\n\nv1.0.1 | 2026-04-25T09:59:57.761Z | user\n\nFull English localization (v1.0.1)\n\nv1.0.0 | 2026-04-25T09:45:45.643Z | user\n\n首次发布\n\nArchive index:\n\nArchive v1.0.4: 5 files, 6477 bytes\n\nFiles: _meta.json (127b), references/amber_parameter_guide.md (498b), references/cpptraj_analysis_guide.md (449b), skill-card.md (2316b), SKILL.md (11119b)\n\nFile v1.0.4:SKILL.md\n\n---\nname: amber-md\ndescription: |\n  Official-style Amber24 molecular dynamics workflow guide for proteins. Includes a standard end-to-end Amber MD procedure, command templates, input-file templates, and a complete manual worked example from PDB download through RMSD/RMSF analysis. This is a documentation-oriented scientific skill for reproducible research workflows in Amber.\nmetadata:\n  {\n    \"openclaw\":\n      {\n        \"emoji\": \"🧬\",\n        \"examples\":\n          [\n            \"how do I run Amber MD on a protein?\",\n            \"show me a standard Amber workflow with commands\",\n            \"give me a complete manual example for Amber RMSD and RMSF analysis\"\n          ]\n      }\n  }\n---\n\n# Amber Molecular Dynamics Simulation\n\n## Overview\n\nThis skill is designed as an **official-style documentation page** for running a standard protein molecular dynamics workflow with **Amber24 / AmberTools24**.\n\nIt is intentionally **documentation-first**:\n- it explains the canonical workflow,\n- provides reusable command templates,\n- provides input-file templates,\n- and gives a full manual example that users can reproduce step by step.\n\nIt does **not** bundle executable automation scripts inside the public ClawHub package.\n\n---\n\n## What this skill helps you do\n\nUse this skill when you need to:\n\n1. prepare a protein structure from the PDB database,\n2. build a solvated Amber system,\n3. run minimization, heating, equilibration, and production MD,\n4. analyze the trajectory with `cpptraj`,\n5. understand what outputs to expect and how to judge simulation quality.\n\n---\n\n## Recommended Amber tools\n\n| Tool | Main purpose |\n|---|---|\n| `pdb4amber` | preprocess PDB structures for Amber |\n| `tleap` | build topology and coordinates |\n| `pmemd.cuda` | GPU production MD |\n| `pmemd` | CPU fallback |\n| `cpptraj` | trajectory analysis |\n| `antechamber` | ligand parameterization when needed |\n| `parmchk2` | missing force-field terms for ligands |\n\n---\n\n## Standard workflow summary\n\n| Stage | Goal | Main output |\n|---|---|---|\n| 1. Structure preparation | clean and standardize input structure | processed PDB |\n| 2. System building | add force field, solvent, ions | `prmtop`, `inpcrd` |\n| 3. Minimization | remove bad contacts | minimized restart file |\n| 4. NVT heating | raise temperature to target value | heated restart + trajectory |\n| 5. NPT equilibration | stabilize density and pressure | equilibrated restart + trajectory |\n| 6. Production MD | generate scientific trajectory | production trajectory |\n| 7. Analysis | compute RMSD/RMSF and related metrics | data tables and plots |\n\n---\n\n## Recommended directory layout\n\n```text\nproject/\n├── input/\n│   └── protein.pdb\n├── build/\n├── md/\n├── analysis/\n└── logs/\n```\n\nA simpler flat directory also works, but a structured layout improves reproducibility.\n\n---\n\n## 1. Structure preparation\n\n### Recommended rules\n\n- If the PDB entry contains multiple NMR models, start with **Model 1**.\n- Remove unsupported ligands if you do not have parameters for them.\n- Keep the protein-only workflow as the default baseline.\n- Use `pdb4amber` to standardize residue names and protonation-related formatting.\n\n### Command template\n\n```bash\nmkdir -p input build md analysis logs\ncd project\n\n# download from RCSB\nwget -O input/1AKI.pdb https://files.rcsb.org/download/1AKI.pdb\n\n# preprocess\npdb4amber -i input/1AKI.pdb -o input/1AKI_amber.pdb --reduce > logs/pdb4amber.log 2>&1\n```\n\n### Notes\n\n- If preprocessing fails because of nonstandard residues, inspect the structure first.\n- For protein-only tutorials, removing unsupported ligands is often the most robust starting point.\n\n---\n\n## 2. System building with tleap\n\n### Recommended setup\n\n- Protein force field: `ff19SB`\n- Water model: `OPC`\n- Box type: truncated octahedron\n- Padding: ~15 Å\n- Neutralization: `Na+` / `Cl-`\n\n### `tleap` input template\n\nSave as `build/tleap.in`:\n\n```bash\nsource leaprc.protein.ff19SB\nsource leaprc.water.opc\n\nmol = loadPDB ../input/1AKI_amber.pdb\n\ndesc mol\naddions mol Cl- 0\naddions mol Na+ 0\nsolvateOct mol OPCBOX 15.0\naddions2 mol Na+ 0\naddions2 mol Cl- 0\n\nsaveAmberParm mol prmtop inpcrd\nsavePDB mol solvated.pdb\nquit\n```\n\n### Run command\n\n```bash\ncd build\ntleap -f tleap.in > ../logs/tleap.log 2>&1\n```\n\n### Expected outputs\n\n- `build/prmtop`\n- `build/inpcrd`\n- `build/solvated.pdb`\n\n---\n\n## 3. Energy minimization\n\nA common two-stage minimization is sufficient for many small-to-medium protein systems.\n\n### Stage 1 minimization template\n\nSave as `md/min1.in`:\n\n```bash\nStage 1 minimization\n &cntrl\n  imin=1,\n  maxcyc=5000,\n  ncyc=2500,\n  ntb=1,\n  ntr=0,\n  cut=10.0,\n  ntpr=500,\n /\n```\n\nRun:\n\n```bash\ncd md\npmemd.cuda -O \\\n  -i min1.in \\\n  -o min1.out \\\n  -p ../build/prmtop \\\n  -c ../build/inpcrd \\\n  -r min1.rst7\n```\n\n### Stage 2 minimization template\n\nSave as `md/min2.in`:\n\n```bash\nStage 2 minimization\n &cntrl\n  imin=1,\n  maxcyc=10000,\n  ncyc=5000,\n  ntb=1,\n  ntr=0,\n  cut=10.0,\n  ntpr=1000,\n /\n```\n\nRun:\n\n```bash\npmemd.cuda -O \\\n  -i min2.in \\\n  -o min2.out \\\n  -p ../build/prmtop \\\n  -c min1.rst7 \\\n  -r min2.rst7\n```\n\n---\n\n## 4. NVT heating\n\n### Heating template\n\nSave as `md/heat.in`:\n\n```bash\nNVT heating\n &cntrl\n  imin=0,\n  irest=0,\n  ntx=1,\n  nstlim=50000,\n  dt=0.002,\n  ntf=2,\n  ntc=2,\n  tempi=0.0,\n  temp0=300.0,\n  ntt=3,\n  gamma_ln=1.0,\n  ntb=1,\n  cut=10.0,\n  ntpr=5000,\n  ntwx=5000,\n /\n```\n\nRun:\n\n```bash\npmemd.cuda -O \\\n  -i heat.in \\\n  -o heat.out \\\n  -p ../build/prmtop \\\n  -c min2.rst7 \\\n  -r heat.rst7 \\\n  -x heat.nc\n```\n\n---\n\n## 5. NPT equilibration\n\n### Equilibration template\n\nSave as `md/equil.in`:\n\n```bash\nNPT equilibration\n &cntrl\n  imin=0,\n  irest=1,\n  ntx=5,\n  nstlim=100000,\n  dt=0.002,\n  ntf=2,\n  ntc=2,\n  temp0=300.0,\n  ntt=3,\n  gamma_ln=1.0,\n  ntb=2,\n  ntp=1,\n  pres0=1.0,\n  cut=10.0,\n  ntpr=10000,\n  ntwx=10000,\n /\n```\n\nRun:\n\n```bash\npmemd.cuda -O \\\n  -i equil.in \\\n  -o equil.out \\\n  -p ../build/prmtop \\\n  -c heat.rst7 \\\n  -r equil.rst7 \\\n  -x equil.nc\n```\n\n---\n\n## 6. Production MD\n\n### 1 ns production template\n\nSave as `md/prod.in`:\n\n```bash\nProduction MD\n &cntrl\n  imin=0,\n  irest=1,\n  ntx=5,\n  nstlim=500000,\n  dt=0.002,\n  ntf=2,\n  ntc=2,\n  temp0=300.0,\n  ntt=3,\n  gamma_ln=1.0,\n  ntb=2,\n  ntp=1,\n  pres0=1.0,\n  cut=10.0,\n  iwrap=1,\n  ntpr=25000,\n  ntwx=12500,\n /\n```\n\nRun:\n\n```bash\npmemd.cuda -O \\\n  -i prod.in \\\n  -o prod.out \\\n  -p ../build/prmtop \\\n  -c equil.rst7 \\\n  -r prod.rst7 \\\n  -x prod.nc\n```\n\n### Useful duration reference\n\n| Target time | `nstlim` with `dt=0.002 ps` |\n|---|---|\n| 1 ns | 500000 |\n| 10 ns | 5000000 |\n| 100 ns | 50000000 |\n\n---\n\n## 7. Trajectory analysis with cpptraj\n\n### Recommended analysis procedure\n\nBefore RMSD/RMSF:\n- strip solvent and ions if you want protein-only metrics,\n- apply `autoimage`,\n- use a consistent atom mask.\n\n### `cpptraj` template\n\nSave as `analysis/analyze.cpptraj`:\n\n```bash\nparm ../build/prmtop\ntrajin ../md/prod.nc 1 last 1\n\nstrip :WAT\nstrip :Na+\nstrip :Cl-\nautoimage\n\nrms out rmsd_ca.dat\natomicfluct out rmsf_ca.dat\nrun\n```\n\nRun:\n\n```bash\ncd analysis\ncpptraj -i analyze.cpptraj > ../logs/cpptraj.log 2>&1\n```\n\n### Typical outputs\n\n- `analysis/rmsd_ca.dat`\n- `analysis/rmsf_ca.dat`\n- `logs/cpptraj.log`\n\n---\n\n## Full manual example: protein-only 1 ns Amber MD\n\nThis example shows a minimal, reproducible manual workflow using **1AKI**.\n\n### Step 1 — create directories\n\n```bash\nmkdir -p amber_1aki/{input,build,md,analysis,logs}\ncd amber_1aki\n```\n\n### Step 2 — download and preprocess structure\n\n```bash\nwget -O input/1AKI.pdb https://files.rcsb.org/download/1AKI.pdb\npdb4amber -i input/1AKI.pdb -o input/1AKI_amber.pdb --reduce > logs/pdb4amber.log 2>&1\n```\n\n### Step 3 — create `build/tleap.in`\n\n```bash\ncat > build/tleap.in << 'EOF'\nsource leaprc.protein.ff19SB\nsource leaprc.water.opc\nmol = loadPDB ../input/1AKI_amber.pdb\naddions mol Cl- 0\naddions mol Na+ 0\nsolvateOct mol OPCBOX 15.0\naddions2 mol Na+ 0\naddions2 mol Cl- 0\nsaveAmberParm mol prmtop inpcrd\nsavePDB mol solvated.pdb\nquit\nEOF\n```\n\nRun:\n\n```bash\ncd build\ntleap -f tleap.in > ../logs/tleap.log 2>&1\ncd ..\n```\n\n### Step 4 — create minimization, heating, equilibration, and production input files\n\nUse the exact templates from the sections above:\n- `md/min1.in`\n- `md/min2.in`\n- `md/heat.in`\n- `md/equil.in`\n- `md/prod.in`\n\n### Step 5 — run MD\n\n```bash\ncd md\npmemd.cuda -O -i min1.in  -o min1.out  -p ../build/prmtop -c ../build/inpcrd -r min1.rst7\npmemd.cuda -O -i min2.in  -o min2.out  -p ../build/prmtop -c min1.rst7       -r min2.rst7\npmemd.cuda -O -i heat.in  -o heat.out  -p ../build/prmtop -c min2.rst7       -r heat.rst7  -x heat.nc\npmemd.cuda -O -i equil.in -o equil.out -p ../build/prmtop -c heat.rst7       -r equil.rst7 -x equil.nc\npmemd.cuda -O -i prod.in  -o prod.out  -p ../build/prmtop -c equil.rst7      -r prod.rst7  -x prod.nc\ncd ..\n```\n\n### Step 6 — analyze trajectory\n\n```bash\ncat > analysis/analyze.cpptraj << 'EOF'\nparm ../build/prmtop\ntrajin ../md/prod.nc 1 last 1\nstrip :WAT\nstrip :Na+\nstrip :Cl-\nautoimage\nrms out rmsd_ca.dat\natomicfluct out rmsf_ca.dat\nrun\nEOF\n\ncd analysis\ncpptraj -i analyze.cpptraj > ../logs/cpptraj.log 2>&1\ncd ..\n```\n\n### Step 7 — inspect results\n\n```bash\nls -lh md/prod.nc analysis/rmsd_ca.dat analysis/rmsf_ca.dat\n```\n\nExpected scientific interpretation:\n- RMSD reaches a stable plateau after equilibration,\n- RMSF is higher at loops and termini,\n- temperature remains close to 300 K,\n- pressure fluctuates around 1 atm during NPT stages.\n\n---\n\n## Common quality checks\n\n| Check | What to look for |\n|---|---|\n| Minimization | energy decreases and no severe bad contacts remain |\n| Heating | temperature ramps smoothly toward 300 K |\n| Equilibration | density/pressure behavior stabilizes |\n| Production RMSD | plateau rather than monotonic drift |\n| RMSF | flexible regions match structural expectation |\n\n---\n\n## Common pitfalls\n\n1. **Using all atoms including water for RMSD**\n   - this often produces meaningless large RMSD values.\n\n2. **Ignoring ligands with missing parameters**\n   - remove them or parameterize them first.\n\n3. **Using inconsistent atom masks**\n   - reference and trajectory selections must match.\n\n4. **Skipping `autoimage`**\n   - periodic boundary effects can distort analysis.\n\n5. **Starting with too large a target simulation**\n   - validate the workflow with 1 ns before 100 ns.\n\n---\n\n## Ligands and nonstandard residues\n\nIf your structure contains:\n- a ligand,\n- a cofactor,\n- a metal center,\n- a nucleotide analog,\n- or another nonstandard residue,\n\nthen you may need:\n- `antechamber`,\n- `parmchk2`,\n- additional `tleap` libraries,\n- or specialized workflows such as `MCPB.py`.\n\nFor a first-pass reproducible workflow, a **protein-only simulation** is often the most robust baseline.\n\n---\n\n## References included in this skill\n\n- `references/amber_parameter_guide.md`\n- `references/cpptraj_analysis_guide.md`\n\n---\n\n## Intended use\n\nThis public ClawHub version is intended for:\n- scientific education,\n- workflow standardization,\n- project planning,\n- manual reproducible Amber execution.\n\nIt is a **public documentation edition** rather than a bundled executable automation package.\n\nFile v1.0.4:_meta.json\n\n{\n  \"ownerId\": \"kn76km32bz2x4rjp91mb223h5982sd5q\",\n  \"slug\": \"amber-md\",\n  \"version\": \"1.0.4\",\n  \"publishedAt\": 1777112694717\n}\n\nFile v1.0.4:references/amber_parameter_guide.md\n\n# Amber MD Parameter Guide\n\n## Core controls\n- `nstlim`: number of MD steps\n- `dt`: timestep in ps\n- `temp0`: target temperature\n- `ntb`: box / boundary mode\n- `ntp`: pressure coupling\n- `cut`: nonbonded cutoff\n- `ntwx`: trajectory write interval\n- `ntpr`: energy print interval\n\n## Common durations\n- 1 ns: `nstlim=500000` with `dt=0.002`\n- 10 ns: `nstlim=5000000`\n- 100 ns: `nstlim=50000000`\n\n## Typical production settings\n- Temperature: 300 K\n- Pressure: 1 atm\n- Timestep: 2 fs\n- Ensemble: NPT\n\nFile v1.0.4:references/cpptraj_analysis_guide.md\n\n# cpptraj Analysis Guide\n\n## Recommended analysis order\n1. Load topology\n2. Load trajectory\n3. Strip water and ions if appropriate\n4. Apply `autoimage`\n5. Compute RMSD\n6. Compute RMSF\n\n## Typical outputs\n- RMSD time series\n- RMSF per residue\n- Radius of gyration\n- Hydrogen-bond summaries\n\n## Common pitfalls\n- Using all atoms including solvent for RMSD\n- Inconsistent masks between reference and trajectory\n- Forgetting periodic imaging correction\n\nFile v1.0.4:skill-card.md\n\n## Description:\n\nOfficial-style Amber24 molecular dynamics workflow guide for proteins, including a standard end-to-end procedure, command templates, input-file templates, and a manual worked example from PDB download through RMSD/RMSF analysis.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[ning-kun](https://clawhub.ai/user/ning-kun)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nScientists, educators, and technical users use this skill to plan and manually run reproducible Amber protein molecular dynamics workflows, from structure preparation through trajectory analysis. It provides workflow guidance, command templates, and quality checks rather than executable automation.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Amber simulations can consume substantial compute or GPU time and create local output files.\n\nMitigation: Review commands before execution, validate with the documented 1 ns workflow first, and run in an appropriate project directory with expected storage and compute capacity.\n\nRisk: Scientific results can be misleading if input structures, ligands, atom masks, or imaging choices are inappropriate.\n\nMitigation: Follow the documented quality checks, remove or parameterize unsupported ligands, use consistent atom masks, and apply trajectory imaging correction before interpreting RMSD or RMSF.\n\n## Reference(s):\n\n- [Amber Parameter Guide](references/amber_parameter_guide.md)\n- [cpptraj Analysis Guide](references/cpptraj_analysis_guide.md)\n- [1AKI PDB Download](https://files.rcsb.org/download/1AKI.pdb)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown guidance with inline bash commands and Amber input-file templates]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Documentation-only output; users review and run commands manually in their Amber environment.]\n\n## Skill Version(s):\n\n1.0.4 (source: server release evidence and artifact metadata)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v1.0.3: 4 files, 3233 bytes\n\nFiles: _meta.json (127b), references/amber_parameter_guide.md (498b), references/cpptraj_analysis_guide.md (449b), SKILL.md (4100b)\n\nFile v1.0.3:SKILL.md\n\n---\nname: amber-md\ndescription: |\n  Amber24 molecular dynamics workflow guide for proteins. This skill explains how to retrieve a PDB structure, prepare a solvated Amber system, run minimization and equilibration, perform production MD, and analyze trajectories with RMSD and RMSF. It is a documentation-oriented scientific skill intended for research workflows.\nmetadata:\n  {\n    \"openclaw\":\n      {\n        \"emoji\": \"🧬\",\n        \"examples\":\n          [\n            \"how do I run Amber MD on a protein?\",\n            \"show me the standard Amber workflow for a PDB structure\",\n            \"how to analyze RMSD and RMSF after Amber simulation?\"\n          ]\n      }\n  }\n---\n\n# Amber Molecular Dynamics Simulation\n\n## Purpose\n\nThis skill provides a **documentation-first Amber MD workflow** for protein simulations.\nIt does **not** execute code by itself. Instead, it gives a clear, reproducible scientific procedure for users working in an Amber24 / AmberTools24 environment.\n\n## Scope\n\nSupported workflow:\n1. Obtain a protein structure from the PDB database\n2. Preprocess the structure for Amber\n3. Build a solvated system with ions\n4. Run minimization\n5. Run NVT heating\n6. Run NPT equilibration\n7. Run production MD\n8. Analyze RMSD and RMSF using cpptraj\n\n## Recommended Amber Components\n\n| Tool | Purpose |\n|---|---|\n| `pdb4amber` | PDB preprocessing |\n| `tleap` | topology and coordinate generation |\n| `pmemd.cuda` | GPU production MD |\n| `pmemd` | CPU fallback |\n| `cpptraj` | trajectory analysis |\n| `antechamber` | ligand parameterization when needed |\n\n## Standard Workflow\n\n### 1. Structure preparation\n- Download a PDB structure from RCSB PDB\n- If the entry is an NMR structure with multiple models, use **Model 1** by default\n- Remove unsupported nonstandard molecules if ligand parameterization is not available\n- Run `pdb4amber` to standardize residue names and add hydrogens\n\n### 2. System building\nRecommended setup:\n- Protein force field: `ff19SB`\n- Water model: `OPC`\n- Neutralization: add `Na+` or `Cl-`\n- Solvent box: truncated octahedron with ~15 Å padding\n\nExpected outputs:\n- `prmtop`\n- `inpcrd`\n- solvated coordinate file\n\n### 3. Energy minimization\nA common two-stage strategy:\n- Stage 1: 5,000 steps\n- Stage 2: 10,000 steps\n\nTypical parameters:\n- `imin=1`\n- `cut=10.0`\n- `ntb=1`\n\n### 4. NVT heating\nTypical setup:\n- 0 → 300 K\n- 100 ps\n- `dt=0.002 ps`\n- Langevin thermostat\n\n### 5. NPT equilibration\nTypical setup:\n- 300 K\n- 1 atm\n- 200 ps\n- constant pressure periodic box\n\n### 6. Production MD\nA common starting point:\n- 1 ns production run\n- `nstlim=500000`\n- `dt=0.002 ps`\n- NPT ensemble\n\nLonger runs:\n- 10 ns: `nstlim=5000000`\n- 100 ns: `nstlim=50000000`\n\n## Trajectory analysis\n\n### RMSD\nRecommended metric:\n- backbone or CA RMSD versus the initial frame\n\n### RMSF\nRecommended metric:\n- per-residue or per-CA fluctuation\n\n### Best practice before analysis\n- strip water molecules and ions\n- apply periodic imaging correction (`autoimage`)\n- use a consistent atom selection mask\n\n## Quality checks\n\n| Metric | Typical healthy range |\n|---|---|\n| CA RMSD | below ~3 Å for a stable small protein |\n| CA RMSF | commonly ~0.3–2.0 Å |\n| Temperature | near target temperature |\n| Pressure | fluctuates around 1 atm |\n\n## Output interpretation\n\n- Stable RMSD plateau: structure is equilibrated\n- High RMSF peaks: flexible loops or termini\n- Large RMSD drift: possible unfolding, poor equilibration, or analysis mask issue\n\n## Notes on ligands\n\nIf a PDB entry contains a ligand, cofactor, or nucleotide analog:\n- standard protein preparation may fail\n- ligand parameterization may be required\n- if parameters are unavailable, protein-only simulation is a valid fallback for many exploratory tasks\n\n## References included in this skill\n\nSee the `references/` directory for:\n- Amber parameter notes\n- cpptraj analysis notes\n\n## Intended use\n\nThis skill is intended for:\n- scientific education\n- workflow standardization\n- Amber project planning\n- interpreting Amber simulation outputs\n\nIt is not intended to hide execution behavior or automate arbitrary external actions.\n\nFile v1.0.3:_meta.json\n\n{\n  \"ownerId\": \"kn76km32bz2x4rjp91mb223h5982sd5q\",\n  \"slug\": \"amber-md\",\n  \"version\": \"1.0.3\",\n  \"publishedAt\": 1777112035856\n}\n\nFile v1.0.3:references/amber_parameter_guide.md\n\n# Amber MD Parameter Guide\n\n## Core controls\n- `nstlim`: number of MD steps\n- `dt`: timestep in ps\n- `temp0`: target temperature\n- `ntb`: box / boundary mode\n- `ntp`: pressure coupling\n- `cut`: nonbonded cutoff\n- `ntwx`: trajectory write interval\n- `ntpr`: energy print interval\n\n## Common durations\n- 1 ns: `nstlim=500000` with `dt=0.002`\n- 10 ns: `nstlim=5000000`\n- 100 ns: `nstlim=50000000`\n\n## Typical production settings\n- Temperature: 300 K\n- Pressure: 1 atm\n- Timestep: 2 fs\n- Ensemble: NPT\n\nFile v1.0.3:references/cpptraj_analysis_guide.md\n\n# cpptraj Analysis Guide\n\n## Recommended analysis order\n1. Load topology\n2. Load trajectory\n3. Strip water and ions if appropriate\n4. Apply `autoimage`\n5. Compute RMSD\n6. Compute RMSF\n\n## Typical outputs\n- RMSD time series\n- RMSF per residue\n- Radius of gyration\n- Hydrogen-bond summaries\n\n## Common pitfalls\n- Using all atoms including solvent for RMSD\n- Inconsistent masks between reference and trajectory\n- Forgetting periodic imaging correction\n\nArchive v1.0.2: 14 files, 14190 bytes\n\nFiles: _meta.json (127b), references/amber_parameter_guide.md (1923b), references/cpptraj_analysis_guide.md (1403b), scripts/extract_nmr_model.py (1221b), scripts/plot_rmsd.py (879b), scripts/plot_rmsf.py (1084b), scripts/run_analysis.sh (2504b), scripts/step1_build.sh (4151b), scripts/step2_min1.sh (1426b), scripts/step3_min2.sh (1251b), scripts/step4_heat.sh (1431b), scripts/step5_equil.sh (1357b), scripts/step6_production.sh (2384b), SKILL.md (6322b)\n\nFile v1.0.2:SKILL.md\n\n---\nname: amber-md\ndescription: |\n  Amber24 molecular dynamics simulation pipeline for proteins. Downloads a PDB structure from RCSB (or uses a local file), builds a solvated system with OPC water and ions, runs energy minimization and NVT/NPT equilibration, then performs production MD under NPT conditions at 300 K. Outputs RMSD and RMSF plots via cpptraj. This is a scientific simulation tool built on the validated Amber MD software suite.\n  Trigger: user requests a molecular dynamics simulation, MD run, or \"run MD on [protein name/PDB ID]\".\nmetadata:\n  {\n    \"openclaw\":\n      {\n        \"emoji\": \"🧬\",\n        \"examples\":\n          [\n            \"run MD on 1AKI with Amber\",\n            \"do a 100ns molecular dynamics simulation\",\n            \"download 8QDK and run Amber MD\",\n            \"plot RMSD and RMSF for my MD trajectory\"\n          ]\n      }\n  }\n---\n\n# Amber Molecular Dynamics Simulation Skill\n\n## Overview\n\nThis skill automates the complete **Amber MD pipeline**: from PDB retrieval to trajectory analysis, no manual steps required. All scripts have been validated on real systems.\n\n### Prerequisites\n\n- **Amber24 + AmberTools24** installed\n- **GPU** strongly recommended (`pmemd.cuda`); CPU fallback available\n\n### Core Modules Used\n\n| Module | Purpose |\n|---|---|\n| `pdb4amber` | PDB preprocessing (protonation, renumbering) |\n| `tleap` | System builder (solvation, ions) |\n| `pmemd.cuda` | GPU-accelerated MD engine |\n| `cpptraj` | Trajectory analysis (RMSD/RMSF/Rg) |\n\n---\n\n## Workflow\n\n### Step 1 — PDB Retrieval & Preprocessing\n\n**Script**: `scripts/step1_build.sh`\n\n```\nInput : PDB ID (e.g. 1AKI) or local .pdb file\nProcess:\n  1. Download from RCSB or copy local file\n  2. Extract NMR Model 1 if multi-model structure\n  3. pdb4amber: protonation, chain renumbering\n  4. tleap: OPC water box + Na+/Cl- neutralization\nOutput: prmtop, inpcrd, solvated.pdb\n```\n\n> ⚠️ **Ligand handling**: Non-standard ligands (cofactors, nucleotide analogs like GMPPNP) require separate parameterization via `antechamber` or `MCPB.py`. If parameterization fails, the skill falls back to **protein-only** simulation.\n\n---\n\n### Step 2 — Energy Minimization\n\n**Scripts**: `scripts/step2_min1.sh` (Stage 1, 5 000 steps) + `scripts/step3_min2.sh` (Stage 2, 10 000 steps)\n\n```\nInput : prmtop + inpcrd\nProcess: Two-stage minimization (solvent relaxed first, then full system)\nOutput: min2.rst7\n```\n\n---\n\n### Step 3 — NVT Heating\n\n**Script**: `scripts/step4_heat.sh`\n\n```\nInput : min2.rst7\nParams: 0 → 300 K, 100 ps, dt = 2 fs\nOutput: heat.rst7 + heat.nc\n```\n\n---\n\n### Step 4 — NPT Equilibration\n\n**Script**: `scripts/step5_equil.sh`\n\n```\nInput : heat.rst7\nParams: 300 K, 1 atm, 200 ps, Langevin thermostat\nOutput: equil.rst7 + equil.nc\n```\n\n---\n\n### Step 5 — Production MD\n\n**Script**: `scripts/step6_production.sh`\n\n```\nInput : equil.rst7\nParams: 300 K, 1 atm, NPT, default 1 ns (500 000 steps @ dt=2 fs)\nOutput: prod.nc (trajectory) + prod.rst7 (final coordinates)\n```\n\n> 💡 **Extend simulation**: change `nstlim` from `500000` (1 ns) → `50000000` (100 ns).\n\n---\n\n### Step 6 — Trajectory Analysis\n\n**Script**: `scripts/run_analysis.sh`\n\n```\nInput : prod.nc\nProcess (cpptraj):\n  1. strip :WAT :Na+ :Cl-\n  2. autoimage\n  3. rms (CA backbone RMSD vs initial frame)\n  4. atomicfluct (CA RMSF)\nOutput:\n  rmsd_ca.dat  - RMSD per frame (Å)\n  rmsf_ca.dat  - RMSF per atom (Å)\n```\n\n---\n\n## Parameter Quick Reference\n\n| Parameter | Meaning | Common Value |\n|---|---|---|\n| `nstlim` | Total steps | 500 000 (1 ns) / 50 000 000 (100 ns) |\n| `dt` | Time step (ps) | 0.002 (2 fs) |\n| `temp0` | Target temperature (K) | 300 |\n| `ntb=2, ntp=1` | NPT ensemble (with pressure coupling) | — |\n| `cut` | Non-bonded cutoff (Å) | 10.0 |\n| `ntwx` | Trajectory write frequency (frames per ns) | 12 500 (≈50 frames/ns) |\n\n### Simulation Duration & Frame Count\n\n```\nframes = nstlim / ntwx\n1 ns  → nstlim=500 000,   ntwx=12 500 → 40 frames\n100 ns → nstlim=50 000 000, ntwx=125 000 → 400 frames\n```\n\n---\n\n## Quality Benchmarks\n\n| Metric | Healthy Range | Interpretation |\n|---|---|---|\n| RMSD (CA) | < 3.0 Å | Conformational stability |\n| RMSF (CA) | 0.3 – 2.0 Å | Atomic fluctuation |\n| Temperature | 298–302 K | NVT/NPT stability |\n| Pressure | 0.8–1.2 atm | NPT equilibrium |\n| Energy | Converged, no drift | Minimization quality |\n\n---\n\n## Output Files\n\n```\nworkdir/\n├── prmtop / inpcrd           # Topology + initial coordinates\n├── min1.rst7 / min2.rst7    # Minimization results\n├── heat.nc / equil.nc        # Heating & equilibration trajectories\n├── prod.nc                   # Production MD trajectory (main result)\n├── prod.rst7                 # Final coordinates\n├── rmsd_ca.dat               # RMSD data (plot with Python/matplotlib)\n├── rmsf_ca.dat               # RMSF data (plot with Python/matplotlib)\n├── rmsd.png / rmsf.png        # Publication-ready figures\n└── cpptraj.log              # Analysis log\n```\n\n---\n\n## Running the Full Pipeline\n\n```bash\ncd /home/user/my_project/\nbash step1_build.sh          # System setup (PDB → tleap)\nbash step2_min1.sh          # Minimization stage 1\nbash step3_min2.sh          # Minimization stage 2\nbash step4_heat.sh          # NVT heating\nbash step5_equil.sh         # NPT equilibration\nbash step6_production.sh    # Production MD (GPU, ~8 min/ns)\nbash run_analysis.sh        # RMSD + RMSF analysis\n```\n\n**Override simulation length (e.g. 100 ns):**\n```bash\nNSTLIM=50000000 bash step6_production.sh\n```\n\n---\n\n## Python Plotting\n\n```bash\n# Plot RMSD\npython3 scripts/plot_rmsd.py rmsd_ca.dat\n# Output: rmsd.png\n\n# Plot RMSF\npython3 scripts/plot_rmsf.py rmsf_ca.dat\n# Output: rmsf.png\n```\n\n---\n\n## FAQ\n\n**Q: The ligand fails to parameterize. What can I do?**\nA: Remove the ligand and simulate the protein only (this is the safest default). For full complex simulation, parameterize the ligand separately with `antechamber` + `parmchk2` + `tleap` using GAFF2 (see `references/antechamber_guide.md`).\n\n**Q: How do I know the system is equilibrated?**\nA: RMSD stabilizes (fluctuation < 0.5 Å over the last half) and temperature/pressure oscillate around setpoints.\n\n**Q: GPU not available?**\nA: Replace `pmemd.cuda` with `pmemd` (CPU-only) in all scripts. Expect 10–50× slower performance.\n\nFile v1.0.2:_meta.json\n\n{\n  \"ownerId\": \"kn76km32bz2x4rjp91mb223h5982sd5q\",\n  \"slug\": \"amber-md\",\n  \"version\": \"1.0.2\",\n  \"publishedAt\": 1777111781213\n}\n\nFile v1.0.2:references/amber_parameter_guide.md\n\n# Amber MD Parameter Quick Reference\n\n## Production MD (NPT)\n\n```\n &cntrl\n  imin=0,          ! 0 = MD mode (1 = minimization)\n  irest=1,         ! 1 = restart (continue from .rst7), 0 = fresh start\n  ntx=5,           ! Read coordinates + velocities (5 = with velocities)\n  nstlim=500000,   ! Total steps (1 ns @ dt=0.002 = 500 000)\n  dt=0.002,        ! Time step (ps), default 2 fs\n  ntf=2,           ! Force evaluation: 2 = SHAKE on H\n  ntc=2,           ! SHAKE constraint on H\n  temp0=300.0,     ! Target temperature (K)\n  ntt=3,           ! Temperature coupling: 3 = Langevin\n  gamma_ln=1.0,   ! Langevin collision frequency (ps^-1)\n  ntb=2,           ! 2 = NPT (periodic boundary, constant pressure)\n  ntp=1,           ! 1 = isobaric (constant pressure)\n  pres0=1.0,       ! Reference pressure (atm)\n  cut=10.0,        ! Non-bonded cutoff (A)\n  iwrap=1,         ! 1 = wrap coordinates back into box\n  ntpr=25000,      ! Print energy every ntpr steps\n  ntwx=12500,     ! Write trajectory frame every ntwx steps\n /\n```\n\n## Energy Minimization\n\n```\n &cntrl\n  imin=1,          ! 1 = minimization mode\n  maxcyc=10000,    ! Maximum minimization steps\n  ncyc=5000,       ! First ncyc steps: steepest descent; rest: conjugate gradient\n  ntb=1,           ! Constant volume\n  ntr=0,           ! 0 = no restraints, 1 = positional restraints\n  cut=10.0,\n  ntpr=500,\n /\n```\n\n## Frame Count Calculation\n\n```\nframes = nstlim / ntwx\n1 ns  → 500 000 / 12 500 = 40 frames\n10 ns → 5 000 000 / 50 000 = 100 frames\n100 ns → 50 000 000 / 125 000 = 400 frames\n```\n\n## Amber Executables\n\n```\n$AMBERHOME/bin/pmemd.cuda   # GPU MD (recommended)\n$AMBERHOME/bin/pmemd        # CPU MD\n$AMBERHOME/bin/sander       # MPI-distributed MD\n$AMBERHOME/bin/tleap        # System builder\n$AMBERHOME/bin/cpptraj      # Trajectory analysis\n$AMBERHOME/bin/pdb4amber    # PDB preprocessor\n$AMBERHOME/bin/antechamber  # Small-molecule parameterization\n```\n\nFile v1.0.2:references/cpptraj_analysis_guide.md\n\n# cpptraj Trajectory Analysis Guide\n\n## Basic Usage\n\n```bash\ncpptraj -p prmtop -i analyze.cpptraj\ncpptraj -p prmtop -i analyze.cpptraj > cpptraj.log 2>&1\n```\n\n## Common Analysis Commands\n\n### RMSD (Backbone CA vs Reference Frame)\n\n```cppraj\nparm prmtop\ntrajin prod.nc 1 last 1\nautoimage\nrms out rmsd_ca.dat\nrun\n```\n\n### RMSF (Atomic Fluctuation)\n\n```cppraj\natomicfluct :1-168@CA out rmsf_ca.dat\n# Output: per-atom RMSF (A)\n```\n\n### Radius of Gyration\n\n```cppraj\nrg :1-168 out rg.dat\n```\n\n### Hydrogen Bond Analysis\n\n```cppraj\nhbond out hbond.dat\n```\n\n### Complete Strip & Analyze\n\n```cppraj\nparm prmtop\ntrajin prod.nc 1 last 1\n\n# Strip solvent and ions\nstrip :WAT\nstrip :Na+\nstrip :Cl-\n\n# Periodic boundary correction\nautoimage\n\n# Analyses\nrms out rmsd_ca.dat\natomicfluct :1-168@CA out rmsf_ca.dat\nrg :1-168 out rg.dat\nhbond out hbond.dat\n\nrun\n```\n\n## Frame Range Syntax\n\n| Syntax | Meaning |\n|---|---|\n| `trajin prod.nc` | All frames |\n| `trajin prod.nc 1 40 1` | Explicit range 1–40 (recommended for reliability) |\n| `trajin prod.nc 1 last 1` | All frames with explicit stride |\n\n> ⚠️ `last` may behave inconsistently across cpptraj versions. Using explicit ranges (`1 40 1`) is safest.\n\n## Python Plotting\n\nStandalone plotting scripts are included in `scripts/`:\n\n```bash\npython3 scripts/plot_rmsd.py rmsd_ca.dat    # → rmsd.png\npython3 scripts/plot_rmsf.py rmsf_ca.dat    # → rmsf.png\n```\n\nArchive v1.0.1: 11 files, 13214 bytes\n\nFiles: _meta.json (127b), references/amber_parameter_guide.md (1923b), references/cpptraj_analysis_guide.md (2019b), scripts/run_analysis.sh (2567b), scripts/step1_build.sh (4763b), scripts/step2_min1.sh (1426b), scripts/step3_min2.sh (1251b), scripts/step4_heat.sh (1431b), scripts/step5_equil.sh (1357b), scripts/step6_production.sh (2399b), SKILL.md (6821b)\n\nFile v1.0.1:SKILL.md\n\n---\nname: amber-md\ndescription: |\n  Amber Molecular Dynamics Simulation Skill. Fully automated pipeline: PDB download → system preparation → energy minimization → NVT/NPT equilibration → production MD → trajectory analysis (RMSD/RMSF). Works with any PDB ID or local PDB file, default 1 ns NPT production, outputs publication-ready RMSD/RMSF plots.\n  Trigger when the user asks about molecular dynamics simulations, MD, or \"run a simulation of protein X\".\nmetadata:\n  {\n    \"openclaw\":\n      {\n        \"emoji\": \"🧬\",\n        \"examples\":\n          [\n            \"run MD on 1AKI with Amber\",\n            \"do a 100ns molecular dynamics simulation\",\n            \"download 8QDK and run Amber MD\",\n            \"plot RMSD and RMSF for my MD trajectory\"\n          ]\n      }\n  }\n---\n\n# Amber Molecular Dynamics Simulation Skill\n\n## Overview\n\nThis skill automates the complete **Amber MD pipeline**: from PDB retrieval to trajectory analysis, no manual steps required. All scripts have been validated on real systems.\n\n### Prerequisites\n\n- **Amber24 + AmberTools24** installed\n- **GPU** strongly recommended (`pmemd.cuda`); CPU fallback available\n\n### Core Modules Used\n\n| Module | Purpose |\n|---|---|\n| `pdb4amber` | PDB preprocessing (protonation, renumbering) |\n| `tleap` | System builder (solvation, ions) |\n| `pmemd.cuda` | GPU-accelerated MD engine |\n| `cpptraj` | Trajectory analysis (RMSD/RMSF/Rg) |\n\n---\n\n## Workflow\n\n### Step 1 — PDB Retrieval & Preprocessing\n\n**Script**: `scripts/step1_build.sh`\n\n```\nInput : PDB ID (e.g. 1AKI) or local .pdb file\nProcess:\n  1. Download from RCSB or copy local file\n  2. Extract NMR Model 1 if multi-model structure\n  3. pdb4amber: protonation, chain renumbering\n  4. tleap: OPC water box + Na+/Cl- neutralization\nOutput: prmtop, inpcrd, solvated.pdb\n```\n\n> ⚠️ **Ligand handling**: Non-standard ligands (cofactors, nucleotide analogs like GMPPNP) require separate parameterization via `antechamber` or `MCPB.py`. If parameterization fails, the skill falls back to **protein-only** simulation.\n\n---\n\n### Step 2 — Energy Minimization\n\n**Scripts**: `scripts/step2_min1.sh` (Stage 1, 5 000 steps) + `scripts/step3_min2.sh` (Stage 2, 10 000 steps)\n\n```\nInput : prmtop + inpcrd\nProcess: Two-stage minimization (solvent relaxed first, then full system)\nOutput: min2.rst7\n```\n\n---\n\n### Step 3 — NVT Heating\n\n**Script**: `scripts/step4_heat.sh`\n\n```\nInput : min2.rst7\nParams: 0 → 300 K, 100 ps, dt = 2 fs\nOutput: heat.rst7 + heat.nc\n```\n\n---\n\n### Step 4 — NPT Equilibration\n\n**Script**: `scripts/step5_equil.sh`\n\n```\nInput : heat.rst7\nParams: 300 K, 1 atm, 200 ps, Langevin thermostat\nOutput: equil.rst7 + equil.nc\n```\n\n---\n\n### Step 5 — Production MD\n\n**Script**: `scripts/step6_production.sh`\n\n```\nInput : equil.rst7\nParams: 300 K, 1 atm, NPT, default 1 ns (500 000 steps @ dt=2 fs)\nOutput: prod.nc (trajectory) + prod.rst7 (final coordinates)\n```\n\n> 💡 **Extend simulation**: change `nstlim` from `500000` (1 ns) → `50000000` (100 ns).\n\n---\n\n### Step 6 — Trajectory Analysis\n\n**Script**: `scripts/run_analysis.sh`\n\n```\nInput : prod.nc\nProcess (cpptraj):\n  1. strip :WAT :Na+ :Cl-\n  2. autoimage\n  3. rms (CA backbone RMSD vs initial frame)\n  4. atomicfluct (CA RMSF)\nOutput:\n  rmsd_ca.dat  - RMSD per frame (Å)\n  rmsf_ca.dat  - RMSF per atom (Å)\n```\n\n---\n\n## Parameter Quick Reference\n\n| Parameter | Meaning | Common Value |\n|---|---|---|\n| `nstlim` | Total steps | 500 000 (1 ns) / 50 000 000 (100 ns) |\n| `dt` | Time step (ps) | 0.002 (2 fs) |\n| `temp0` | Target temperature (K) | 300 |\n| `ntb=2, ntp=1` | NPT ensemble (with pressure coupling) | — |\n| `cut` | Non-bonded cutoff (Å) | 10.0 |\n| `ntwx` | Trajectory write frequency (frames per ns) | 12 500 (≈50 frames/ns) |\n\n### Simulation Duration & Frame Count\n\n```\nframes = nstlim / ntwx\n1 ns  → nstlim=500 000,   ntwx=12 500 → 40 frames\n100 ns → nstlim=50 000 000, ntwx=125 000 → 400 frames\n```\n\n---\n\n## Quality Benchmarks\n\n| Metric | Healthy Range | Interpretation |\n|---|---|---|\n| RMSD (CA) | < 3.0 Å | Conformational stability |\n| RMSF (CA) | 0.3 – 2.0 Å | Atomic fluctuation |\n| Temperature | 298–302 K | NVT/NPT stability |\n| Pressure | 0.8–1.2 atm | NPT equilibrium |\n| Energy | Converged, no drift | Minimization quality |\n\n---\n\n## Output Files\n\n```\nworkdir/\n├── prmtop / inpcrd           # Topology + initial coordinates\n├── min1.rst7 / min2.rst7    # Minimization results\n├── heat.nc / equil.nc        # Heating & equilibration trajectories\n├── prod.nc                   # Production MD trajectory (main result)\n├── prod.rst7                 # Final coordinates\n├── rmsd_ca.dat               # RMSD data (plot with Python/matplotlib)\n├── rmsf_ca.dat               # RMSF data (plot with Python/matplotlib)\n└── cpptraj.log              # Analysis log\n```\n\n---\n\n## Running the Full Pipeline\n\n```bash\ncd /home/user/my_project/\nbash step1_build.sh          # System setup (PDB → tleap)\nbash step2_min1.sh          # Minimization stage 1\nbash step3_min2.sh          # Minimization stage 2\nbash step4_heat.sh          # NVT heating\nbash step5_equil.sh         # NPT equilibration\nbash step6_production.sh    # Production MD (GPU, ~8 min/ns)\nbash run_analysis.sh        # RMSD + RMSF analysis\n```\n\n**Override simulation length (e.g. 100 ns):**\n```bash\nNSTLIM=50000000 bash step6_production.sh\n```\n\n---\n\n## Python Plotting\n\n```python\nimport numpy as np, matplotlib.pyplot as plt\n\n# RMSD\nd = np.loadtxt(\"rmsd_ca.dat\")\nplt.figure(figsize=(8, 5))\nplt.plot(d[:, 0], d[:, 1], 'steelblue', lw=1.5)\nplt.axhline(d[:, 1].mean(), color='r', ls='--', lw=1.2,\n            label=f'mean = {d[:, 1].mean():.3f} A')\nplt.xlabel(\"Frame\"); plt.ylabel(\"RMSD (A)\")\nplt.title(\"CA Backbone RMSD\"); plt.legend(); plt.grid(alpha=0.3)\nplt.savefig(\"rmsd.png\", dpi=150)\n\n# RMSF\nf = np.loadtxt(\"rmsf_ca.dat\", skiprows=1)\nplt.figure(figsize=(8, 5))\nplt.plot(f[:, 0], f[:, 1], 'forestgreen', lw=0.9)\nplt.axhline(f[:, 1].mean(), color='orange', ls='--',\n            label=f'mean = {f[:, 1].mean():.3f} A')\nplt.xlabel(\"Residue\"); plt.ylabel(\"RMSF (A)\")\nplt.title(\"CA Atomic Fluctuation\"); plt.legend(); plt.grid(alpha=0.3)\nplt.savefig(\"rmsf.png\", dpi=150)\n```\n\n---\n\n## FAQ\n\n**Q: The ligand fails to parameterize. What can I do?**\nA: Remove the ligand and simulate the protein only (this is the safest default). For full complex simulation, parameterize the ligand separately with `antechamber` + `parmchk2` + `tleap` using GAFF2 (see `references/antechamber_guide.md`).\n\n**Q: How do I know the system is equilibrated?**\nA: RMSD stabilizes (fluctuation < 0.5 Å over the last half) and temperature/pressure oscillate around setpoints.\n\n**Q: GPU not available?**\nA: Replace `pmemd.cuda` with `pmemd` (CPU-only) in all scripts. Expect 10–50× slower performance.\n\nFile v1.0.1:_meta.json\n\n{\n  \"ownerId\": \"kn76km32bz2x4rjp91mb223h5982sd5q\",\n  \"slug\": \"amber-md\",\n  \"version\": \"1.0.1\",\n  \"publishedAt\": 1777111197761\n}\n\nFile v1.0.1:references/amber_parameter_guide.md\n\n# Amber MD Parameter Quick Reference\n\n## Production MD (NPT)\n\n```\n &cntrl\n  imin=0,          ! 0 = MD mode (1 = minimization)\n  irest=1,         ! 1 = restart (continue from .rst7), 0 = fresh start\n  ntx=5,           ! Read coordinates + velocities (5 = with velocities)\n  nstlim=500000,   ! Total steps (1 ns @ dt=0.002 = 500 000)\n  dt=0.002,        ! Time step (ps), default 2 fs\n  ntf=2,           ! Force evaluation: 2 = SHAKE on H\n  ntc=2,           ! SHAKE constraint on H\n  temp0=300.0,     ! Target temperature (K)\n  ntt=3,           ! Temperature coupling: 3 = Langevin\n  gamma_ln=1.0,   ! Langevin collision frequency (ps^-1)\n  ntb=2,           ! 2 = NPT (periodic boundary, constant pressure)\n  ntp=1,           ! 1 = isobaric (constant pressure)\n  pres0=1.0,       ! Reference pressure (atm)\n  cut=10.0,        ! Non-bonded cutoff (A)\n  iwrap=1,         ! 1 = wrap coordinates back into box\n  ntpr=25000,      ! Print energy every ntpr steps\n  ntwx=12500,     ! Write trajectory frame every ntwx steps\n /\n```\n\n## Energy Minimization\n\n```\n &cntrl\n  imin=1,          ! 1 = minimization mode\n  maxcyc=10000,    ! Maximum minimization steps\n  ncyc=5000,       ! First ncyc steps: steepest descent; rest: conjugate gradient\n  ntb=1,           ! Constant volume\n  ntr=0,           ! 0 = no restraints, 1 = positional restraints\n  cut=10.0,\n  ntpr=500,\n /\n```\n\n## Frame Count Calculation\n\n```\nframes = nstlim / ntwx\n1 ns  → 500 000 / 12 500 = 40 frames\n10 ns → 5 000 000 / 50 000 = 100 frames\n100 ns → 50 000 000 / 125 000 = 400 frames\n```\n\n## Amber Executables\n\n```\n$AMBERHOME/bin/pmemd.cuda   # GPU MD (recommended)\n$AMBERHOME/bin/pmemd        # CPU MD\n$AMBERHOME/bin/sander       # MPI-distributed MD\n$AMBERHOME/bin/tleap        # System builder\n$AMBERHOME/bin/cpptraj      # Trajectory analysis\n$AMBERHOME/bin/pdb4amber    # PDB preprocessor\n$AMBERHOME/bin/antechamber  # Small-molecule parameterization\n```\n\nFile v1.0.1:references/cpptraj_analysis_guide.md\n\n# cpptraj Trajectory Analysis Guide\n\n## Basic Usage\n\n```bash\ncpptraj -p prmtop -i analyze.cpptraj\ncpptraj -p prmtop -i analyze.cpptraj > cpptraj.log 2>&1\n```\n\n## Common Analysis Commands\n\n### RMSD (Backbone CA vs Reference Frame)\n\n```cppraj\nparm prmtop\ntrajin prod.nc 1 last 1\nautoimage\nrms out rmsd_ca.dat\nrun\n```\n\n### RMSF (Atomic Fluctuation)\n\n```cppraj\natomicfluct :1-168@CA out rmsf_ca.dat\n# Output: per-atom RMSF (A)\n```\n\n### Radius of Gyration\n\n```cppraj\nrg :1-168 out rg.dat\n```\n\n### Hydrogen Bond Analysis\n\n```cppraj\nhbond out hbond.dat\n```\n\n### Complete Strip & Analyze\n\n```cppraj\nparm prmtop\ntrajin prod.nc 1 last 1\n\n# Strip solvent and ions\nstrip :WAT\nstrip :Na+\nstrip :Cl-\n\n# Periodic boundary correction\nautoimage\n\n# Analyses\nrms out rmsd_ca.dat\natomicfluct :1-168@CA out rmsf_ca.dat\nrg :1-168 out rg.dat\nhbond out hbond.dat\n\nrun\n```\n\n## Frame Range Syntax\n\n| Syntax | Meaning |\n|---|---|\n| `trajin prod.nc` | All frames |\n| `trajin prod.nc 1 40 1` | Explicit range 1–40 (recommended for reliability) |\n| `trajin prod.nc 1 last 1` | All frames with explicit stride |\n\n> ⚠️ `last` may behave inconsistently across cpptraj versions. Using explicit ranges (`1 40 1`) is safest.\n\n## Python Plotting\n\n```python\nimport numpy as np, matplotlib.pyplot as plt\n\n# RMSD\nd = np.loadtxt(\"rmsd_ca.dat\")\nplt.figure(figsize=(8, 5))\nplt.plot(d[:, 0], d[:, 1], 'steelblue', lw=1.5)\nplt.axhline(d[:, 1].mean(), color='r', ls='--', lw=1.2,\n            label=f'mean = {d[:, 1].mean():.3f} A')\nplt.xlabel(\"Frame\"); plt.ylabel(\"RMSD (A)\")\nplt.title(\"CA Backbone RMSD\"); plt.legend(); plt.grid(alpha=0.3)\nplt.savefig(\"rmsd.png\", dpi=150)\n\n# RMSF\nf = np.loadtxt(\"rmsf_ca.dat\", skiprows=1)\nplt.figure(figsize=(8, 5))\nplt.plot(f[:, 0], f[:, 1], 'forestgreen', lw=0.9)\nplt.axhline(f[:, 1].mean(), color='orange', ls='--',\n            label=f'mean = {f[:, 1].mean():.3f} A')\nplt.xlabel(\"Residue\"); plt.ylabel(\"RMSF (A)\")\nplt.title(\"CA Atomic Fluctuation\"); plt.legend(); plt.grid(alpha=0.3)\nplt.savefig(\"rmsf.png\", dpi=150)\n```\n\nArchive v1.0.0: 11 files, 13597 bytes\n\nFiles: _meta.json (127b), references/amber_parameter_guide.md (1665b), references/cpptraj_analysis_guide.md (1915b), scripts/run_analysis.sh (2565b), scripts/step1_build.sh (5269b), scripts/step2_min1.sh (1426b), scripts/step3_min2.sh (1251b), scripts/step4_heat.sh (1431b), scripts/step5_equil.sh (1357b), scripts/step6_production.sh (2399b), SKILL.md (6506b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: amber-md\ndescription: |\n  Amber 分子动力学模拟技能。自动完成从 PDB 下载 → 体系构建 → 能量最小化 → NVT/NPT 平衡 → 成品 MD → 轨迹分析（RMSD/RMSF）全流程。支持任意 PDB 结构，默认 1 ns NPT 成品模拟，输出可发表的 RMSD/RMSF 图。\n  当用户需要进行分子动力学模拟、描述\"A/B 蛋白的 MD\"或\"做一下动力学研究\"时触发。\nmetadata:\n  {\n    \"openclaw\":\n      {\n        \"emoji\": \"🧬\",\n        \"examples\":\n          [\n            \"对 1AKI 做 Amber MD\",\n            \"跑一个 100ns 的分子动力学\",\n            \"下载 8QDK 做动力学模拟\",\n            \"给我画一下 RMSD 和 RMSF 图\"\n          ]\n      }\n  }\n---\n\n# Amber 分子动力学模拟技能\n\n## Overview\n\n本技能提供从 **PDB 结构获取 → 体系预处理 → 能量最小化 → NVT/NPT 平衡 → 成品 MD → 轨迹分析** 的全流程自动化。无需手动操作，所有脚本均经过实际运行验证。\n\n### 适用场景\n\n- 蛋白质（或含配体/金属中心的蛋白）MD 模拟\n- 配体-蛋白复合物的动力学稳定性评估\n- RMSD / RMSF / Rg / 氢键等标准分析\n- 不同 PDB 结构之间的 MD 对比研究\n\n### 依赖环境\n\n本技能运行在已安装 **Amber24 + AmberTools24** 的环境中，核心模块：\n\n| 模块 | 用途 |\n|---|---|\n| `pdb4amber` | PDB 预处理（加氢、重编号） |\n| `tleap` | 体系构建（溶胶、加离子） |\n| `pmemd.cuda` | GPU 加速 MD 引擎 |\n| `cpptraj` | 轨迹分析（RMSD/RMSF 等） |\n\n---\n\n## 工作流程\n\n### 步骤 1：PDB 获取与预处理\n\n**脚本**：`scripts/step1_build.sh`\n\n```\n输入：任意 PDB ID（如 1AKI）或本地 PDB 文件\n处理：\n  1. 从 RCSB 下载 PDB 文件\n  2. pdb4amber 自动加氢、重编号\n  3. 处理非标准残基（配体需额外参数化）\n  4. tleap 构建溶剂盒子（OPC 水模型）+ 中和离子\n输出：prmtop（拓扑文件）、inpcrd（坐标文件）、solvated.pdb\n```\n\n> ⚠️ **配体处理**：标准 GAFF2 无法处理的非标准配体（如 GMPPNP、辅酶因子等）需先用 `antechamber` 或 `MCPB.py` 单独参数化。若无法参数化，技能将自动回退到**仅蛋白主体**模拟。\n\n---\n\n### 步骤 2：能量最小化\n\n**脚本**：`scripts/step2_min1.sh`（阶段1，全体系 5000 步）\n\n**脚本**：`scripts/step3_min2.sh`（阶段2，全体系 10000 步）\n\n```\n输入：prmtop + inpcrd\n处理：两阶段能量最小化（溶剂先放松，再全体系）\n输出：min2.rst7（最终坐标）\n```\n\n---\n\n### 步骤 3：NVT 升温\n\n**脚本**：`scripts/step4_heat.sh`\n\n```\n输入：min2.rst7\n参数：0 → 300 K，100 ps，dt = 2 fs\n输出：heat.rst7 + heat.nc\n```\n\n---\n\n### 步骤 4：NPT 平衡\n\n**脚本**：`scripts/step5_equil.sh`\n\n```\n输入：heat.rst7\n参数：300 K，1 atm，200 ps，dt = 2 fs\n输出：equil.rst7 + equil.nc\n```\n\n---\n\n### 步骤 5：成品 MD\n\n**脚本**：`scripts/step6_production.sh`\n\n```\n输入：equil.rst7\n参数：300 K，1 atm，NPT 系综，默认 1 ns（500000 步 @ dt=2 fs）\n输出：prod.nc（轨迹）+ prod.rst7（最终坐标）\n```\n\n> 💡 **扩展时长**：将 `nstlim` 从 `500000`（1 ns）改为 `50000000`（100 ns）即可。\n\n---\n\n### 步骤 6：轨迹分析\n\n**脚本**：`scripts/run_analysis.sh`\n\n```\n输入：prod.nc\n处理（cpptraj）：\n  1. strip :WAT :Na+ :Cl-（去溶剂/离子）\n  2. autoimage（周期性边界校正）\n  3. rms（CA backbone RMSD vs 初始帧）\n  4. atomicfluct（CA RMSF）\n输出：\n  rmsd_ca.dat  - 每帧 RMSD（Å）\n  rmsf_ca.dat  - 每原子 RMSF（Å）\n```\n\n---\n\n## 核心参数速查\n\n| 参数 | 含义 | 常用值 |\n|---|---|---|\n| `nstlim` | 总步数 | 500000（1 ns）/ 50000000（100 ns）|\n| `dt` | 时间步长（ps） | 0.002（2 fs）|\n| `temp0` | 目标温度（K） | 300 |\n| `ntb=2, ntp=1` | NPT 系综（含压力耦合） | — |\n| `cut` | 非键截断（Å） | 10.0 |\n| `ntwx` | 轨迹输出频率 | 每 25000 步（50 帧/ns）|\n\n---\n\n## 质量评估标准\n\n| 指标 | 健康范围 | 含义 |\n|---|---|---|\n| RMSD（CA） | < 3.0 Å | 构象稳定性 |\n| RMSF（CA） | 0.3 – 2.0 Å | 原子涨落 |\n| 温度 | 298–302 K | NVT/NPT 稳定性 |\n| 压力 | 0.8–1.2 atm | NPT 平衡 |\n| 能量 | 收敛无漂移 | 最小化质量 |\n\n---\n\n## 输出产物\n\n```\nworkdir/\n├── prmtop / inpcrd         # 拓扑 + 初始坐标\n├── min1.rst7 / min2.rst7   # 极小化结果\n├── heat.nc / equil.nc      # 升温和平衡轨迹\n├── prod.nc                 # 成品 MD 轨迹（主要结果）\n├── prod.rst7               # 成品 MD 最终坐标\n├── rmsd_ca.dat             # RMSD 数据（可绘图）\n├── rmsf_ca.dat             # RMSF 数据（可绘图）\n└── cpptraj.log             # 分析日志\n```\n\n---\n\n## 脚本调用方式\n\n```bash\n# 完整流程（自动依次执行）\nbash step1_build.sh\nbash step2_min1.sh\nbash step3_min2.sh\nbash step4_heat.sh\nbash step5_equil.sh\nbash step6_production.sh   # GPU，约 8 分钟 / ns\nbash run_analysis.sh       # RMSD + RMSF\n\n# 扩展到 100 ns\nsed -i 's/nstlim=500000/nstlim=50000000/' step6_production.sh\nbash step6_production.sh\n```\n\n---\n\n## Python 绘图示例\n\n```python\nimport numpy as np, matplotlib.pyplot as plt\n\n# RMSD\nd = np.loadtxt(\"rmsd_ca.dat\")\nplt.figure(figsize=(8,5))\nplt.plot(d[:,0], d[:,1], 'b-', linewidth=1.5)\nplt.axhline(d[:,1].mean(), color='r', ls='--', label=f'mean={d[:,1].mean():.3f} Å')\nplt.xlabel(\"Frame\"); plt.ylabel(\"RMSD (Å)\")\nplt.title(\"CA Backbone RMSD\"); plt.legend(); plt.grid(alpha=0.3)\nplt.savefig(\"rmsd.png\", dpi=150)\n\n# RMSF\nf = np.loadtxt(\"rmsf_ca.dat\", skiprows=1)\nplt.figure(figsize=(8,5))\nplt.plot(f[:,0], f[:,1], 'g-', linewidth=0.8)\nplt.axhline(f[:,1].mean(), color='orange', ls='--', label='mean')\nplt.xlabel(\"Residue\"); plt.ylabel(\"RMSF (Å)\")\nplt.title(\"CA Atomic Fluctuation\"); plt.legend(); plt.grid(alpha=0.3)\nplt.savefig(\"rmsf.png\", dpi=150)\n```\n\n---\n\n## 常见问题\n\n**Q: 配体无法处理怎么办？**\nA: 去掉配体只模拟蛋白主体是最稳妥方案。如需含配体，可使用 `antechamber` + `parmchk2` + `tleap` 对配体单独进行 GAFF2 参数化（参考 references/antechamber_guide.md）。\n\n**Q: 如何判断体系是否平衡？**\nA: RMSD 在后期（> 500 ps）趋于稳定且波动 < 0.5 Å，温度/压力在设定值附近小幅波动即视为平衡。\n\n**Q: GPU 不可用怎么办？**\nA: 将 `pmemd.cuda` 替换为 `pmemd`（CPU 版本），速度约慢 10–50 倍。\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn76km32bz2x4rjp91mb223h5982sd5q\",\n  \"slug\": \"amber-md\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1777110345643\n}\n\nFile v1.0.0:references/amber_parameter_guide.md\n\n# Amber MD 参数速查指南\n\n## MD 输入文件（.in）参数详解\n\n### 成品 MD（NPT）\n\n```\n &cntrl\n  imin=0,         ! 非最小化（0=MD，1=最小化）\n  irest=1,       ! 1=重启（续跑），0=从头开始\n  ntx=5,         ! 读取坐标+速度（5=含速度）\n  nstlim=500000, ! 总步数（1 ns = 500000 @ dt=0.002）\n  dt=0.002,      ! 时间步长（ps），通常 0.002\n  ntf=2,         ! 约束力计算（2=SHAKE on H）\n  ntc=2,         ! SHAKE 约束（2=on H）\n  temp0=300.0,   ! 目标温度（K）\n  ntt=3,         ! 温度耦合：3=Langevin\n  gamma_ln=1.0,  ! Langevin 碰撞频率\n  ntb=2,         ! 2=NPT（有压强耦合）\n  ntp=1,         ! 1=等压（恒压）\n  pres0=1.0,     ! 参考压力（atm）\n  cut=10.0,      ! 非键截断（Å）\n  iwrap=1,       ! 1=坐标回盒\n  ntpr=25000,    ! 每 ntpr 步输出能量\n  ntwx=12500,    ! 每 ntwx 步输出轨迹帧\n /\n```\n\n### 能量最小化\n\n```\n &cntrl\n  imin=1,        ! 1=最小化模式\n  maxcyc=10000,  ! 最大步数\n  ncyc=5000,     ! 前 ncyc 步用最陡下降，之后用共轭梯度\n  ntb=1,         ! 固定体积\n  ntr=0,         ! 0=无位置约束，1=有约束\n  cut=10.0,\n  ntpr=500,\n /\n```\n\n## 轨迹帧数计算\n\n```\n帧数 = nstlim / ntwx\n示例：nstlim=500000, ntwx=12500 → 40 帧（1 ns）\n```\n\n## Amber 可执行文件路径\n\n```\n$AMBERHOME/bin/pmemd.cuda   # GPU MD（推荐）\n$AMBERHOME/bin/pmemd        # CPU MD\n$AMBERHOME/bin/sander       # MPI 分布式 MD\n$AMBERHOME/bin/tleap        # 体系构建\n$AMBERHOME/bin/cpptraj      # 轨迹分析\n$AMBERHOME/bin/pdb4amber    # PDB 预处理\n$AMBERHOME/bin/antechamber  # 小分子参数化\n```\n\nFile v1.0.0:references/cpptraj_analysis_guide.md\n\n# cpptraj 轨迹分析指南\n\n## 基础用法\n\n```bash\n# 标准调用\ncpptraj -p prmtop -i analyze.cpptraj\n\n# 带日志输出\ncpptraj -p prmtop -i analyze.cpptraj > cpptraj.log 2>&1\n```\n\n## 常用分析命令\n\n### RMSD（骨架原子 vs 参考帧）\n\n```cppraj\nparm prmtop\ntrajin prod.nc 1 last 1\nautoimage\nrms out rmsd_ca.dat\nrun\n```\n\n### RMSF（原子涨落）\n\n```cppraj\natomicfluct :1-168@CA out rmsf_ca.dat\n# 输出：每原子 RMSF（Å）\n```\n\n### 回旋半径（Rg）\n\n```cppraj\nrg :1-168 out rg.dat\n```\n\n### 氢键分析\n\n```cppraj\nhbond out hbond.dat\n```\n\n### 去水 + 全部分析\n\n```cppraj\nparm prmtop\ntrajin prod.nc 1 last 1\n\n# 去溶剂和离子\nstrip :WAT\nstrip :Na+\nstrip :Cl-\n\n# 周期性边界校正\nautoimage\n\n# 各项分析\nrms out rmsd_ca.dat\natomicfluct :1-168@CA out rmsf_ca.dat\nrg :1-168 out rg.dat\nhbond out hbond.dat\n\nrun\n```\n\n## 帧范围说明\n\n| 写法 | 含义 |\n|---|---|\n| `trajin prod.nc` | 全部帧 |\n| `trajin prod.nc 1 40 1` | 显式指定 1–40 帧（部分版本 `last` 有 bug）|\n| `trajin prod.nc 1 last 1` | 全部帧（`last` 需显式指定间隔）|\n\n## 绘图（Python）\n\n```python\nimport numpy as np, matplotlib.pyplot as plt\n\n# RMSD\nd = np.loadtxt(\"rmsd_ca.dat\")\nplt.figure(figsize=(8,5))\nplt.plot(d[:,0], d[:,1], 'steelblue', lw=1.5)\nplt.axhline(d[:,1].mean(), color='r', ls='--', lw=1.2,\n            label=f'mean = {d[:,1].mean():.3f} A')\nplt.xlabel(\"Frame\"); plt.ylabel(\"RMSD (A)\")\nplt.title(\"CA Backbone RMSD\"); plt.legend(); plt.grid(alpha=0.3)\nplt.savefig(\"rmsd.png\", dpi=150)\n\n# RMSF\nf = np.loadtxt(\"rmsf_ca.dat\", skiprows=1)\nplt.figure(figsize=(8,5))\nplt.plot(f[:,0], f[:,1], 'forestgreen', lw=0.8)\nplt.axhline(f[:,1].mean(), color='orange', ls='--',\n            label=f'mean = {f[:,1].mean():.3f} A')\nplt.xlabel(\"Residue\"); plt.ylabel(\"RMSF (A)\")\nplt.title(\"CA Atomic Fluctuation\"); plt.legend(); plt.grid(alpha=0.3)\nplt.savefig(\"rmsf.png\", dpi=150)\n```","readmeExcerpt":"Skill: Amber Molecular Dynamics Simulation Owner: ning-kun Summary: Official-style Amber24 molecular dynamics workflow guide for proteins. Includes a standard end-to-end Amber MD procedure, command templates, input-file templ... Tags: amber:1.0.4, latest:1.0.4, md:1.0.4, molecular-dynamics:1.0.4, simulation:1.0.4 Version history: v1.0.4 | 2026-04-25T10:24:54.717Z | user v1.0.4: upgraded public documentation into an o","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"project/\n├── input/\n│   └── protein.pdb\n├── build/\n├── md/\n├── analysis/\n└── logs/"},{"language":"bash","snippet":"mkdir -p input build md analysis logs\ncd project\n\n# download from RCSB\nwget -O input/1AKI.pdb https://files.rcsb.org/download/1AKI.pdb\n\n# preprocess\npdb4amber -i input/1AKI.pdb -o input/1AKI_amber.pdb --reduce > logs/pdb4amber.log 2>&1"},{"language":"bash","snippet":"source leaprc.protein.ff19SB\nsource leaprc.water.opc\n\nmol = loadPDB ../input/1AKI_amber.pdb\n\ndesc mol\naddions mol Cl- 0\naddions mol Na+ 0\nsolvateOct mol OPCBOX 15.0\naddions2 mol Na+ 0\naddions2 mol Cl- 0\n\nsaveAmberParm mol prmtop inpcrd\nsavePDB mol solvated.pdb\nquit"},{"language":"bash","snippet":"cd build\ntleap -f tleap.in > ../logs/tleap.log 2>&1"},{"language":"bash","snippet":"Stage 1 minimization\n &cntrl\n  imin=1,\n  maxcyc=5000,\n  ncyc=2500,\n  ntb=1,\n  ntr=0,\n  cut=10.0,\n  ntpr=500,\n /"},{"language":"bash","snippet":"cd md\npmemd.cuda -O \\\n  -i min1.in \\\n  -o min1.out \\\n  -p ../build/prmtop \\\n  -c ../build/inpcrd \\\n  -r min1.rst7"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: amber-md\ndescription: |\n  Official-style Amber24 molecular dynamics workflow guide for proteins. Includes a standard end-to-end Amber MD procedure, command templates, input-file templates, and a complete manual worked example from PDB download through RMSD/RMSF analysis. This is a documentation-oriented scientific skill for reproducible research workflows in Amber.\nmetadata:\n  {\n    \"openclaw\":\n      {\n        \"emoji\": \"🧬\",\n        \"examples\":\n          [\n            \"how do I run Amber MD on a protein?\",\n            \"show me a standard Amber workflow with commands\",\n            \"give me a complete manual example for Amber RMSD and RMSF analysis\"\n          ]\n      }\n  }\n---\n\n# Amber Molecular Dynamics Simulation\n\n## Overview\n\nThis skill is designed as an **official-style documentation page** for running a standard protein molecular dynamics workflow with **Amber24 / AmberTools24**.\n\nIt is intentionally **documentation-first**:\n- it explains the canonical workflow,\n- provides reusable command templates,\n- provides input-file templates,\n- and gives a full manual example that users can reproduce step by step.\n\nIt does **not** bundle executable automation scripts inside the public ClawHub package.\n\n---\n\n## What this skill helps you do\n\nUse this skill when you need to:\n\n1. prepare a protein structure from the PDB database,\n2. build a solvated Amber system,\n3. run minimization, heating, equilibration, and production MD,\n4. analyze the trajectory with `cpptraj`,\n5. understand what outputs to expect and how to judge simulation quality.\n\n---\n\n## Recommended Amber tools\n\n| Tool | Main purpose |\n|---|---|\n| `pdb4amber` | preprocess PDB structures for Amber |\n| `tleap` | build topology and coordinates |\n| `pmemd.cuda` | GPU production MD |\n| `pmemd` | CPU fallback |\n| `cpptraj` | trajectory analysis |\n| `antechamber` | ligand parameterization when needed |\n| `parmchk2` | missing force-field terms for ligands |\n\n---\n\n## Standard workflow summary\n\n| Stage | Goal | Main output |\n|---|---|---|\n| 1. Structure preparation | clean and standardize input structure | processed PDB |\n| 2. System building | add force field, solvent, ions | `prmtop`, `inpcrd` |\n| 3. Minimization | remove bad contacts | minimized restart file |\n| 4. NVT heating | raise temperature to target value | heated restart + trajectory |\n| 5. NPT equilibration | stabilize density and pressure | equilibrated restart + trajectory |\n| 6. Production MD | generate scientific trajectory | production trajectory |\n| 7. Analysis | compute RMSD/RMSF and related metrics | data tables and plots |\n\n---\n\n## Recommended directory layout\n\n```text\nproject/\n├── input/\n│   └── protein.pdb\n├── build/\n├── md/\n├── analysis/\n└── logs/\n```\n\nA simpler flat directory also works, but a structured layout improves reproducibility.\n\n---\n\n## 1. Structure preparation\n\n### Recommended rules\n\n- If the PDB entry contains multiple NMR models, start with **Model 1**.\n- Remove unsupported ligands if you do not have parameter"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn76km32bz2x4rjp91mb223h5982sd5q\",\n  \"slug\": \"amber-md\",\n  \"version\": \"1.0.4\",\n  \"publishedAt\": 1777112694717\n}"},{"path":"references/amber_parameter_guide.md","content":"# Amber MD Parameter Guide\n\n## Core controls\n- `nstlim`: number of MD steps\n- `dt`: timestep in ps\n- `temp0`: target temperature\n- `ntb`: box / boundary mode\n- `ntp`: pressure coupling\n- `cut`: nonbonded cutoff\n- `ntwx`: trajectory write interval\n- `ntpr`: energy print interval\n\n## Common durations\n- 1 ns: `nstlim=500000` with `dt=0.002`\n- 10 ns: `nstlim=5000000`\n- 100 ns: `nstlim=50000000`\n\n## Typical production settings\n- Temperature: 300 K\n- Pressure: 1 atm\n- Timestep: 2 fs\n- Ensemble: NPT"},{"path":"references/cpptraj_analysis_guide.md","content":"# cpptraj Analysis Guide\n\n## Recommended analysis order\n1. Load topology\n2. Load trajectory\n3. Strip water and ions if appropriate\n4. Apply `autoimage`\n5. Compute RMSD\n6. Compute RMSF\n\n## Typical outputs\n- RMSD time series\n- RMSF per residue\n- Radius of gyration\n- Hydrogen-bond summaries\n\n## Common pitfalls\n- Using all atoms including solvent for RMSD\n- Inconsistent masks between reference and trajectory\n- Forgetting periodic imaging correction"},{"path":"skill-card.md","content":"## Description:\n\nOfficial-style Amber24 molecular dynamics workflow guide for proteins, including a standard end-to-end procedure, command templates, input-file templates, and a manual worked example from PDB download through RMSD/RMSF analysis.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[ning-kun](https://clawhub.ai/user/ning-kun)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nScientists, educators, and technical users use this skill to plan and manually run reproducible Amber protein molecular dynamics workflows, from structure preparation through trajectory analysis. It provides workflow guidance, command templates, and quality checks rather than executable automation.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Amber simulations can consume substantial compute or GPU time and create local output files.\n\nMitigation: Review commands before execution, validate with the documented 1 ns workflow first, and run in an appropriate project directory with expected storage and compute capacity.\n\nRisk: Scientific results can be misleading if input structures, ligands, atom masks, or imaging choices are inappropriate.\n\nMitigation: Follow the documented quality checks, remove or parameterize unsupported ligands, use consistent atom masks, and apply trajectory imaging correction before interpreting RMSD or RMSF.\n\n## Reference(s):\n\n- [Amber Parameter Guide](references/amber_parameter_guide.md)\n- [cpptraj Analysis Guide](references/cpptraj_analysis_guide.md)\n- [1AKI PDB Download](https://files.rcsb.org/download/1AKI.pdb)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown guidance with inline bash commands and Amber input-file templates]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Documentation-only output; users review and run commands manually in their Amber environment.]\n\n## Skill Version(s):\n\n1.0.4 (source: server release evidence and artifact metadata)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment."}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"Official-style Amber24 molecular dynamics workflow guide for proteins. Includes a standard end-to-end Amber MD procedure, command templates, input-file templ... Skill: Amber Molecular Dynamics Simulation Owner: ning-kun Summary: Official-style Amber24 molecular dynamics workflow guide for proteins. Includes a standard end-to-end Amber MD procedure, command templates, input-file templ... Tags: amber:1.0.4, latest:1.0.4, md:1.0.4, molecular-dynamics:1.0.4, simulation:1.0.4 Version history: v1.0.4 | 2026-04-25T10:24:54.717Z | user v1.0.4: upgraded public documentation into an o","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1202,"uniquenessScore":52,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-11T20:54:34.055Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-11T20:54:34.055Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-15T03:21:41.443Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}