Skip to main content
Glama

Auto-LAMMPS MCP v1.0

在 Codex 或 DeepSeek Harness 中,用自然语言完成材料模拟研究。

Agent /ˈeɪdʒənt/(智能体)直接编写 LAMMPS、Python 脚本,读取日志并修正;Auto-LAMMPS 负责项目保存、超算执行、后台监视、分析和成果交接。

v1.0 提供本地记忆、验证复用和批量研究:用户可以说“沿用上次方案,再做20次独立重复”。Agent 检索过去工作,已覆盖条件直接准备正式批次,只为变化补充验证。具体安装版本与实测进度见交付记录。下面的固定安装提示词继续作为统一入口。

当前使用环境:macOS · Codex / 官方 DeepSeek Harness(DSH)· Slurm 超算。

下载正式版 v1.0 · Python 包版本 1.0.0。Linux 源码与隔离安装通过自动检查;Windows 原生安装仍在后续交付计划中。

复制安装提示词 · 开始第一项研究 · 查看后台进展 · 版本与进度 · 文档导航

从这里开始

  1. 打开 Codex 或官方 DSH,进入能访问本机文件、执行命令的工作对话。

  2. 复制下方完整安装提示词,粘贴到对话中。Agent 会读取给 Agent 的安装指南,识别环境并推进安装。

  3. 按引导选择研究目录、已有超算连接和计算环境。安装完成后,选择新建研究、继续项目或分析结果。

安装会复用平台已有的模型设置。新电脑先登录所选 Agent 平台;需要提交计算时,配置自己的 Slurm 账号、远端工作目录和 LAMMPS 环境。已有结构、势函数或结果目录可以直接提供给 Agent。

Related MCP server: matmcp

复制给 Agent 的安装提示词

点击下面代码块右上角的复制按钮。这是项目固定安装提示词;后续会话继续使用同一个入口。

请帮我安装和配置 Auto-LAMMPS,让我可以直接在当前对话中提出材料模拟需求,由你调用软件完成计算、监视、分析和结果展示。

项目仓库:https://github.com/XPYY233/auto-lammps-mcp

请先读取仓库首页指向的“给 Agent 的安装指南”,按发布版本对应的说明完成以下工作。

1. 判断当前环境
识别操作系统、当前 Agent 平台,以及你是否能够执行本机命令、安装插件和调用 MCP。
能安装插件时,优先安装包含工作指南和工具配置的插件。
有本机执行能力时,完成本地安装。已经接入服务时,先检查已有连接。
需要我切换会话或操作平台设置时,告诉我具体打开哪里、填写什么,并保存接续信息。

2. 检查已有安装
检查软件版本、真实安装位置、后台服务和平台连接。
区分尚未安装、命令路径未生效、版本需要更新和执行权限不足,分别处理。
复用已有可用安装,保留研究数据和其他软件配置。

3. 完成安装与平台接入
安装当前发布版本、科研工作指南、命令行、MCP 连接和持久后台。
使用安装器返回的真实路径继续工作。
安装后运行环境检查,并从当前 Agent 会话实际调用软件工具,确认我已经可以在这里使用它。

4. 引导我完成基本配置
先发现已有配置,再集中询问仍然缺少的内容:研究项目保存位置;超算连接、账号和计算队列;远端工作目录和已有 LAMMPS 环境;任务资源设置;后台自动接续和通知偏好。
提供推荐值,并用简短中文解释用途。已有配置让我直接选择;密码、密钥和验证码通过本机凭据入口或登录窗口填写。

5. 验证后台与恢复
检查后台服务、任务保存、通知和后台 Codex 接续能力。
保存配置进度,让我关闭对话后能够继续。
需要重新打开会话时,给我一句可以直接复制的接续提示词。

6. 带我完成第一次使用
配置完成后展示实际配置摘要,让我选择:新建研究、继续已有项目、分析已有结果。
如果我选择首次计算验证,先展示计算内容和资源设置。我说“开始”后,完成提交、进度反馈、数据回收、图表和结果文件。

7. 出错时继续处理
根据当前版本的帮助和检查结果定位问题。
告诉我已经完成什么、卡在哪里、下一步是什么。
保留有效配置和已有成果,从失败处恢复。

整个过程使用简短中文。每完成一个阶段告诉我结果;需要我操作时,一次只给当前必需的步骤。
完成后告诉我如何开始任务、查看进度、暂停或停止、查看结果,以及换一个对话继续研究。

安装后,Agent 应分别告诉你:实际软件版本、当前对话连接、研究目录、超算连接、计算环境、后台接续状态,以及下一步。 连接核验会在当前对话中真实调用工具。

安装中断时,在新对话里说:

继续配置 Auto-LAMMPS,从上次已经完成的步骤接着做。

第一次研究怎么开始

你的起点

可以直接对 Agent 说

新建研究

“建立一个 Ni 晶体,在 500 K 下平衡,再进行动力学计算。给我温度、体积、总能、逐原子势能和轨迹。先列出方案与资源。”

继续已有项目

“导入我选中的项目目录,检查结构、势函数、辅助脚本和运行环境,告诉我接下来能做什么。”

分析已有结果

“分析这个目录里的 LAMMPS 日志和轨迹,画出温度、能量随时间的变化,并展示对应结构。”

Agent 会整理研究目标、固定条件和成果清单,先查已有资源,再编写原生脚本。首次计算会展示实际内容和资源设置;你说“开始”后提交到超算。

研究过程中,你可以继续这样操作:

想做什么

对话示例

应看到什么

查看进度

“现在到哪一步了?”

当前阶段、最近证据时间、已完成内容和下一步

修改方案

“保留加载方向,再增加两个温度。”

方案差异、新增计算与受影响成果

排查异常

“这一组为什么没有进展?”

日志依据、最后确认状态与恢复动作

暂停派发

“暂停后续计算。”

新派发已停止;现有作业的实际状态

停止一组

“停止 600 K 这一组,其他继续。”

选中的任务与取消回执

看图与结构

“展示应力突降前后的曲线和结构。”

对应时间点、图表、结构或轨迹文件入口

追加分析

“更换拟合区间,保留原结果。”

复用数据的新分析版本与比较

导出交接

“导出给同门继续使用。”

数据、脚本、图表、轨迹、报告与研究记录

结果通过当前平台的文件预览以及 OVITO 等科研工具打开。原生脚本、结构和数据保存在普通目录中,可以继续编辑和分析。完整用户指南 →

后台如何继续工作

flowchart TD
    U["你:描述需求、查看进度、提出修改"] <--> A["Codex / DSH 中的 Agent"]
    A <--> M["Auto-LAMMPS MCP:原生文件与科研工具"]
    M <--> S["持久任务服务:记录、队列、监视、恢复"]
    S <--> H["Slurm 超算:提交、计算、日志与回收"]
    S <--> R["分析与交付:数据、图表、结构和成果包"]
    S <--> B["后台 Codex:读取证据、诊断与接续"]

后台保存任务进度,按集群合并查询作业,跟进日志和输出变化。需要判断时启动后台 Codex,已经确定的上传、回收、分析和导出由队列推进。

  • 关闭聊天: 电脑在线且后台服务运行时,任务继续监视与处理;重要阶段、异常和需要决定的事项通过已配置的通知提醒。

  • 电脑休眠或断网: 已提交的远端作业继续;本机恢复后先核对实际状态,再接续工作。

  • 暂时不自动处理: 说“这项研究先由当前对话处理,暂停后续后台自动接续”。

  • 回到项目: 说“查看未读进展”,或使用下面的接续句。

进展会围绕以下内容展示:

任务名称|已完成内容|当前阶段|最近证据时间|下一步|需要决定的事项

在同一台电脑的新 Codex 或 DSH 对话中说:

使用 Auto-LAMMPS,继续我最近的 Ni 研究,先告诉我完成了什么、现在到哪里、还有什么没交付。

跨电脑时,先在新电脑使用上面的固定提示词完成安装,再导入研究成果包;完整环境迁移使用备份与恢复流程。后台工作与恢复说明 →

复用已有资源与官方命令说明

已有势函数、作者项目、脚本和已验证运行环境可以登记到研究资源库。Agent 先按材料与计算方法查找,再核对来源、版本、元素映射和实际文件。

接入我已有的资源库,先找 Ni 势函数和对应的 LAMMPS 代码用法,核对已有文件后复用。

软件提供按引擎版本读取、缓存官方命令说明的工具。脚本生成与修正时可读取命令格式、示例和限制;提交准备会检查输入快照中的已知错误。资源检索会区分已有文件和待获取的索引条目。资源与命令使用方法 →

当前版本与进度

入口

当前内容

正式版 v1.0

Python 包 1.0.0;MCP/CLI、持久任务、资源与官方语法、本地记忆、验证复用、批次执行、分析版本和成果交接

P0–P4 开发与阶段检查

功能开发完成;阶段证据见交付记录

Ni 开发流程

已完成 256 原子、500 K 平衡 5 ps、NVE 10 ps;交付五张图、11 帧轨迹、报告与完整研究包,保留三次失败和后续成功记录

当前整体验收

官方 DSH 正在完成 NbTiZrMoV 新研究的完整应用流程;独立安装、真实批次、四类研究矩阵、故障恢复和七天监视继续按计划推进

固定安装提示词默认选择正式发布包;Agent 使用同一发布标签下的指南、skill 和安装器。版本变更见发布说明,实测记录见交付进度、验证记录与验收指标。

手动安装与维护

准备可用的 Git 与 Python。安装器优先通过 uv 创建 Python 3.13 环境;不使用 uv 时,需要 Python 3.12 或更新版本。

在一个新的代码目录中获取固定发布版本:

git clone https://github.com/XPYY233/auto-lammps-mcp.git
cd auto-lammps-mcp
git checkout --detach v1.0

接入 Codex:

python3 scripts/install.py --source . --host codex

接入官方 DSH,把示例路径替换为自己的 Harness 安装目录或 dsh 程序路径:

python3 scripts/install.py --source . --host dsh \
  --dsh-runtime /absolute/path/to/deepseek-harness --dsh-profile web

安装器返回实际命令路径、平台配置、已完成步骤和接续入口。按回执重新加载对话中的工具,然后让 Agent 调用 MCP doctor 与 task_list,核对当前连接和原任务。

发布页同时提供 wheel、安装器和校验摘要,详见给 Agent 的安装指南。

更新与回退: 让 Agent 读取已有安装回执,选择固定版本并保留研究数据;需要回退时恢复上一个版本并核对原任务。操作说明 →

备份与迁移: 保存数据库、项目、快照及交接记录;新电脑恢复后先核对原作业,再接续执行。操作说明 →

卸载: 停止本项目后台、移除本项目 MCP 连接和 skill,再选择保留或迁移研究目录。操作说明 →

文档导航

你需要什么

入口

让 Agent 安装和配置

给 Agent 的安装指南

每一步怎样提需求、看结果

用户指南

关闭聊天、恢复失败、备份迁移

后台与恢复

查看当前完成情况和证据

交付进度 · 验证记录

查看完整产品目标与验收

批准计划 · 验收指标

理解模块和扩展接口

开发说明 · 资源来源

阅读开发与交付手册

发布版 PDF · 可编辑源稿

下载版本或反馈问题

Releases · Issues

参与开发

从明确的用户任务开始,通过 Issue、独立分支、PR 和自动检查完成修改。MCP 与 alm call 连接同一任务服务;alm help 提供当前版本的工具参数。

uv venv --python 3.13
uv pip install -r requirements.lock.txt
uv pip install -e .
.venv/bin/pytest -q
.venv/bin/python scripts/check_public_tree.py
.venv/bin/python scripts/render_review_traceability.py --check

项目采用 Apache-2.0。第三方资源按各自来源和许可使用,参见资产来源记录。

Available Tools

75 tools
analysis_submitA
DestructiveIdempotent

Run a native Python script against existing data in a persistent analysis queue.

script and inputs are project-relative files. They are copied, with subdirectories preserved, into analysis/<operation_id>/. Python's working directory is that analysis directory, so scripts read input paths relative to its root, not their own scripts/ folder. outputs are required nonempty paths relative to the analysis directory. The installed analysis environment provides NumPy and Matplotlib. Example: script='scripts/plot.py', inputs=['data.csv'], outputs=['plot.png','statistics.csv']. Reuse previous results without editing script paths: input_bindings={'data.csv':'analysis//data.csv'} maps workspace paths to project source files. inputs may be [] when bindings are provided. The receipt retains source paths, hashes and each independent input copy. Returns operation_id; release the task lease before polling. operation_get returns output paths, SHA-256 values and analysis.log. Use file_read for logs, artifact_preview for the actual image, then artifact_register to link each result to requirements. For batch analysis set batch_id and batch_files (paths relative to each collected run). The service generates batch-samples.json and samples/<run_id>/... inputs, preserving planned/included/excluded members. The script chooses scientific statistics and reports their denominator. method is optional native-analysis context (purpose, units, window, uncertainty, rationale); supersedes links an earlier analysis operation without deleting its files. Source/version facts are recorded by the service. Physics evaluations use run_prepare/run_submit on HPC.

Inputs: op_id (required), inputs (required), script (required), outputs (required), +8 more. Returns: result

artifact_previewC
Read-onlyIdempotent

Show an existing file with its hash, artifact sources and freshness. MCP embeds supported images directly.

Inputs: path (required), task_id (required).

artifact_registerA
DestructiveIdempotent

Link an actual output to the user's requirement IDs, source files and verification evidence.

path, sources and checks[].evidence are project-relative. The service reads their hashes; do not calculate or repeat hashes in the request. Each check is {'receipt_id': 'check_...'} from check_run/check_get. The receipt must include this artifact and its actual sources. Service-derived measurements determine passed. Historical passed/method/evidence claims are retained as unverified and cannot complete a requirement. The latest registration for this path supplies its current checks and complete requirement_ids. Earlier verdicts remain in history. A failed current check blocks its linked requirements and reopens a completed task; record corrected evidence before re-reviewing. An analysis result path normally starts with analysis/<operation_id>/. Returns the new task revision. description labels the result for users. supersedes explicitly replaces earlier artifact IDs with a revised output, preserving their files/reviews. Replacement must retain their requirement IDs. Other results remain current.

Inputs: path (required), op_id (required), checks (required), sources (required), +6 more. Returns: result

background_requestB
DestructiveIdempotent

Save a continuation intent even while another executor owns this task. after_operations lists this task's queued operations to finish before Codex starts; failed dependencies also wake it for diagnosis. End the model turn after receiving the saved operation ID. expected_outputs contains only concrete project-relative file paths. Task hold/pause still controls dispatch.

Inputs: op_id (required), task_id (required), revision (required), instruction (required), +2 more. Returns: result

backup_createB
DestructiveIdempotent

Queue a verified database/files backup outside the state directory. Release task leases before starting.

Inputs: op_id (required), destination (required). Returns: result

batch_controlB
DestructiveIdempotent

Pause future members, resume pending members, or stop one batch and cancel its active jobs. Other batches retain their controls.

Inputs: op_id (required), action (required), batch_id (required), revision (required). Returns: result

batch_datasetA
Read-onlyIdempotent

Preview batch analysis inputs from collected per-run files. Return independent sample/seed/condition identities, eligible members and all exclusions. files are relative to each run directory. Use analysis_submit(batch_id,batch_files) to generate batch-samples.json and samples/<run_id>/... automatically. Partial failed-run analysis can use ordinary input_bindings with an explicit method.

Inputs: files (required), task_id (required), batch_id (required). Returns: result

batch_getA
Read-onlyIdempotent

Read batch progress, sample identities, controls, costs and latest output steps/time. Missing accounting stays unknown. Computational completion leads to output analysis/checks. Read run_get for full evidence or include_details for all member evidence.

Inputs: batch_id (required), include_details. Returns: result

batch_listB
Read-onlyIdempotent

Find durable batches in a resumed project. Return compact counts, controls, costs and batch_get links; follow next_offset for more.

Inputs: task_id (required), limit, offset. Returns: result

batch_prepareB
DestructiveIdempotent

Prepare independent production repeats using one validated seed-variable template. samples: [{sample_id,seed}]. Uses existing validation, creates immutable per-sample snapshots and no preflight jobs. Call batch_submit to persist dispatch intent.

Inputs: op_id (required), plan_id (required), samples (required), task_id (required), +3 more. Returns: result

batch_submitB
DestructiveIdempotent

Persist all prepared batch members atomically within current task authorization and submission count. Restart/retry retains member identities. Concurrency remains the task policy.

Inputs: op_id (required), batch_id (required), revision (required), token. Returns: result

checker_checkC
DestructiveIdempotent

Check a completed analysis using a qualified program's fixed criteria and exact script version. Queue a service receipt for the measured output and companion artifacts.

Inputs: op_id (required), output (required), task_id (required), revision (required), +4 more. Returns: result

checker_registerB
DestructiveIdempotent

Qualify a native checker after analysis_submit ran independent positive and negative fixtures. cases: [{operation_id,output,expected_passed}]. criteria is one JSON values/bounds check; its path is supplied from each actual output. Freeze script hash, fixtures, expected results and criteria; all cases must match.

Inputs: name (required), cases (required), op_id (required), task_id (required), +3 more. Returns: result

check_getC
Read-onlyIdempotent

Read service-measured values, fixed criteria, coverage and current source integrity.

Inputs: task_id (required), receipt_id (required). Returns: result

check_runB
DestructiveIdempotent

Queue checks of existing data; no physics. Each has kind,path and optional unit,basis,role (required/advisory/candidate). Candidate/advisory results never revoke required evidence. file: min_bytes,contains; csv: columns,min_rows,bounds; json: values,bounds; thermo: phase={fix_id,style,group,marker,timestep}, columns,min_rows,energy_column,max_abs_drift,time_column. Use log_phases facts; block is a legacy selector. Raw energy drift during heat/work exchange needs a task-specific basis and balance=task_specific or an accounted energy column. trajectory: columns,min_frames,atoms; energy_jump: before_block,after_block,column,expected,tolerance. artifacts binds companion files. Read operation_get/check_get.

Inputs: op_id (required), checks (required), task_id (required), revision (required), +2 more. Returns: result

completion_checkC
DestructiveIdempotent

Check original deliverables against non-stale files and recorded scientific checks. Missing outputs keep the task open.

Inputs: op_id (required), task_id (required), revision (required), token. Returns: result

connection_checkB
DestructiveIdempotent

Queue an SSH/Slurm and remote directory check. This does not run a simulation.

Inputs: op_id (required). Returns: result

data_inspectC
Read-onlyIdempotent

Inspect actual LAMMPS thermo or per-atom trajectory data. Columns, frame counts and identities come from files.

Inputs: kind (required), path (required), task_id (required). Returns: result

doctorC
Read-onlyIdempotent

Inspect installation, persisted setup and actual tool/cluster/background receipts separately.

Inputs: no named parameters. Returns: result

environment_captureC
DestructiveIdempotent

Read remote installed engine/launcher bytes and resolved shared libraries; no engine invocation or physics. commands lists actual executables used by the native entry, e.g. ['lmp','mpirun'] or absolute paths. files lists extra absolute plugin/container/runtime dependencies. Returns an operation; use its runtime id in validation_plan. Repeating this capture measures changes without a new preflight job. Failures retain phase/exit_code in error.evidence; diagnose profile/module/file-reading failure from that receipt. Engine startup checks use accounted HPC.

Inputs: op_id (required), task_id (required), commands (required), revision (required), +2 more. Returns: result

environment_verifyC
DestructiveIdempotent

Verify a completed environment_check run using collected engine output and a scientific verification marker.

Inputs: op_id (required), run_id (required), evidence_path (required), expected_text (required). Returns: result

eventsB
Read-onlyIdempotent

Read durable progress since a cursor; optional consumer resumes its unread cursor without marking read.

Inputs: after, limit, task_id, consumer. Returns: result

events_ackC
Destructive

Mark displayed events as read for one platform/session.

Inputs: through (required), consumer (required). Returns: result

file_patchB
DestructiveIdempotent

Edit a single exact fragment; new files use null expected_sha256 and empty old. Returns diff and derived hash.

Inputs: new (required), old (required), path (required), op_id (required), +4 more. Returns: result

file_readB
Read-onlyIdempotent

Read bytes as UTF-8 text with byte offsets, content hash and explicit continuation. Binary files use artifact paths.

Inputs: path (required), task_id (required), limit, offset. Returns: result

helpB
Read-onlyIdempotent

Read one tool's complete signature, or page the generated catalog using next_offset until more is false.

Inputs: limit, method, offset. Returns: result

lammps_input_reviewB
Read-onlyIdempotent

Read an existing native input and report known compute-current mistakes with line numbers, content hash and official references. Read dependencies for dynamic control flow, then verify on accounted HPC. No script rewriting or physics execution.

Inputs: path (required), task_id (required), engine_version. Returns: result

lammps_referenceA
Read-onlyIdempotent

Read cached official Syntax, Examples and Restrictions for the actual engine version, e.g. command='compute pe/atom', engine_version='28Mar2023'. Follow next_offset until more=false. Fetches only the requested official LAMMPS page and retains its source/hash; use before writing or repairing native inputs.

Inputs: command (required), engine_version (required), limit, offset. Returns: result

legacy_inspectA
Read-onlyIdempotent

Inspect a historical SQLite catalog through a read-only connection; never initialize its old controller.

Inputs: database (required). Returns: result

log_phasesC
Destructive

Inspect actual executed LAMMPS phases in an existing log, including fixes, groups, time step, units and normalization. Use those facts as thermo phase selectors; repeated runs/headers need no hand-counted block index. Reads existing data only.

Inputs: path (required), task_id (required). Returns: result

memory_getB
Read-onlyIdempotent

Read one memory and its correction history with source task/event/file references.

Inputs: task_id (required), memory_id (required). Returns: result

memory_rebuildB
DestructiveIdempotent

Re-index retained task events into observed memories. Existing corrections and disabled records are preserved. No historical claim becomes verified.

Inputs: op_id (required), task_id (required). Returns: result

memory_recordB
DestructiveIdempotent

Save a candidate method/failure/preference, correct it, or disable it. Service measurements create verified memories. action: record, correct, disable. Disabling a validation memory also suspends reuse of that plan.

Inputs: kind (required), op_id (required), title (required), summary (required), +8 more. Returns: result

operation_getB
Read-onlyIdempotent

Read work progress/result after any conversation. Export progress reports phase, processed files/input bytes, written archive bytes and observation time; these measure that phase, not scientific completion. Large results preserve IDs and link to operation_read; follow its paths and cursors for every detail. include_details=True returns the full record for local consumers.

Inputs: operation_id (required), include_details. Returns: result

operation_readA
Read-onlyIdempotent

Read lossless pages of a work record. path=['result','files'] reads a snapshot manifest; child paths preserve literal filenames. Objects/arrays return items; long values link to deeper paths. offset is nonnegative; limit/text_limit are positive requested upper bounds, capped at 50 items/4096 characters per page. Text uses character offsets. Follow next_offset while more is true.

Inputs: operation_id (required), path, limit, offset, +1 more. Returns: result

operation_retryC
DestructiveIdempotent

Resume failed/interrupted non-submission work. Creates a separate attempt and preserves the original failure.

Inputs: op_id (required), operation_id (required). Returns: result

project_exportA
DestructiveIdempotent

Export project files, original request, required results, history, rules, job accounting and continuation instructions to ZIP.

Returns operation_id immediately; release the lease before waiting. operation_get returns the archive's absolute path and hash. Another task can pass that ZIP to project_import. The archive is outside project_path, so do not register its absolute path as a project file. By default this satisfies all kind='archive' requirements after verifying the real ZIP. Run completion_check after export completes.

Inputs: op_id (required), task_id (required), revision (required), token, +1 more. Returns: result

project_importA
DestructiveIdempotent

Import a directory or exported ZIP into the task's project_path, retaining relative paths, binary files and helpers.

Existing identical files are reused; differing files cause a conflict, never overwrite. ZIP hashes and original records are checked. Returns operation_id immediately. Release any task_claim before waiting on operation_get; read task_get for the new revision after completion.

Inputs: op_id (required), source (required), task_id (required), revision (required), +1 more. Returns: result

project_optimizeA
DestructiveIdempotent

Queue storage sharing with existing verified snapshots on clone-capable volumes. Every project path, byte content, permission, modification time and attempt remains; original external files and archives stay in place. Files without snapshots or cloning support are skipped. Release the task lease before polling operation_get; the receipt records each path and observed free space.

Inputs: op_id (required), task_id (required), revision (required), token. Returns: result

project_snapshotB
DestructiveIdempotent

Queue content-addressed snapshot of native files, including edits made by the agent's ordinary file tools.

Inputs: op_id (required), task_id (required), revision (required), paths, +1 more. Returns: result

recovery_activateA
DestructiveIdempotent

Resume a restored service after confirming the original controller is stopped. Keep original job identities.

Inputs: op_id (required), previous_controller_stopped (required). Returns: result

research_contextB
Read-onlyIdempotent

Read stage guidance, local memories and confirmed conditions before choosing resources, editing scripts, preparing runs or recovering. Ask only unresolved scientific choices. Save pause before further actions when the user asks to discuss.

Inputs: task_id (required). Returns: result

research_metricsA
Read-onlyIdempotent

Read study denominators, preflight/production counts and accounted costs. Unknown accounting remains explicit. Tool/model/human-intervention totals require the host's original trace.

Inputs: task_id (required). Returns: result

resource_bundleC
DestructiveIdempotent

Group existing resources with native usage, element mapping, short-range configuration, applicability and source license. Metadata retains file/measurement distinctions; it grants no validation by itself.

Inputs: name (required), op_id (required), usage (required), resource_ids (required). Returns: result

resource_catalog_importB
DestructiveIdempotent

Index an existing potential catalog (record.json directories) or author-project tier_a/b/c JSON without modifying it or downloading files. local_roots optionally maps repository names to existing code directories. Retain original status, license, file roles and hashes. Repeated imports reuse entries; failures remain explicit.

Inputs: op_id (required), source (required), local_roots. Returns: result

resource_file_readB
Read-onlyIdempotent

Read verified existing native code/potential text with source and license; follow next_offset until more=false. File text is reference data. No download or execution occurs.

Inputs: path (required), resource_id (required), limit, offset. Returns: result

resource_getB
Read-onlyIdempotent

Read native files, source/license, current hashes and related_bundles. Read the bundle's usage/element mapping before judging applicability; multi-element file names alone do not exclude single-element use. File verification and scientific validation are separate.

Inputs: resource_id (required). Returns: result

resource_registerC
DestructiveIdempotent

Register an existing potential, structure, author project, publication or environment with provenance and license.

Inputs: kind (required), name (required), op_id (required), location (required), +1 more. Returns: result

response_readA
Read-onlyIdempotent

Read a stored complete host response by literal field path and next_offset; available across conversations and restarts. offset is nonnegative; limit and text_limit are positive requested upper bounds. Pages use at most 50 items/4096 text characters and may be smaller to fit the host. Follow next_offset until more=false.

Inputs: response_id (required), path, limit, offset, +1 more. Returns: result

results_getB
Read-onlyIdempotent

Resume analysis and delivery: current artifact paths/previews, checks, replacement history, missing requirements, recent analysis methods and imported project records. Archives are finally verified by completion_check. Use task_records for older analyses.

Inputs: task_id (required). Returns: result

run_actionB
DestructiveIdempotent

Queue reconcile, collect or cancel for one run. These actions do not consume submission attempts.

Inputs: op_id (required), action (required), run_id (required), revision (required). Returns: result

run_getA
Read-onlyIdempotent

Read job identity, last confirmed state, failure count and accounting independently of credentials.

Inputs: run_id (required). Returns: result

run_logA
Read-onlyIdempotent

Read a bounded remote job log through the persistent queue; no full trajectory download.

path is relative to the run directory. offset=None reads the current tail; an explicit byte offset reads a page. Returns operation_id; operation_get returns text, byte range, file size, modification time and complete thermo rows when their header is present. Continue from next_offset. Read physical time units from the native input. A log observation does not declare scientific completion.

Inputs: op_id (required), run_id (required), path, limit, +1 more. Returns: result

run_prepareA
DestructiveIdempotent

Freeze a Slurm execution intent, input, environment and rules. Returns revision for the next mutation. Exact repetition reuses the same run; changed validation_id/stage never silently returns an old binding. To replace a confirmed unsubmitted prepared/cancelled/staging_failed run, use a new intent and supersedes=old run_id, retaining the sample_id. Submitted/unknown runs must be reconciled first. Native entry is a project-relative shell script; read lammps_reference before generation. Literal -in/-i inputs are reviewed before dispatch.

Inputs: entry (required), op_id (required), intent (required), task_id (required), +10 more. Returns: result

run_submitC
DestructiveIdempotent

Persist submission intent and return immediately. Repeated calls return the same run; unknown dispatch reconciles first.

Inputs: op_id (required), run_id (required), revision (required), token. Returns: result

setup_configureA
DestructiveIdempotent

Save shared setup progress. Sections: cluster, background, notifications. Store SSH aliases, never passwords. For one project's background preference use task_background instead of changing this shared setting.

Inputs: op_id (required), values (required), section (required). Returns: result

task_backgroundA
DestructiveIdempotent

Enable or hold future automatic Codex continuations for one project. Running continuations finish their current turn; this does not cancel them or active analyses/jobs. Read active_background before taking over. To interrupt immediately use task_control pause and read current work before resuming. Pending decisions remain saved. Shared setup and other projects stay unchanged.

Inputs: op_id (required), enabled (required), task_id (required), revision (required), +1 more. Returns: result

task_claimB
Destructive

Acquire or renew the single writer lease. An expired token can never write.

Inputs: owner (required), task_id (required), token, seconds. Returns: result

task_controlB
DestructiveIdempotent

Pause future dispatch and queued file-producing work, resume, or stop and cancel active jobs. Persists intent first. Existing remote jobs keep running on pause; log reads, collection and diagnostic checks remain available.

Inputs: op_id (required), action (required), task_id (required), revision (required). Returns: result

task_createC
DestructiveIdempotent

Create a project from original words and deliverables. Each requirement needs id and description; use kind='archive' for an exported ZIP, otherwise kind='file'.

Inputs: op_id (required), title (required), request (required), requirements (required), +3 more. Returns: result

task_getB
Read-onlyIdempotent

Resume a project with original goals, rules, current lease owner and compact progress. executor identifies its owner without sharing write tokens; active_background identifies ongoing automatic recovery even while it awaits queued work. Follow operation_get before starting a competing repair. Use task_records for paged history. include_details=True returns the complete record for export.

Inputs: task_id (required), include_details. Returns: result

task_listA
Read-onlyIdempotent

Find tasks by title or ID. Resolve ambiguous names before changing a task.

Inputs: limit, query. Returns: result

task_recordsC
Read-onlyIdempotent

Read runs, artifacts, work or stages without repeating the project. Large work records link to operation_read for full details. Follow next_offset when more is true.

Inputs: kind (required), task_id (required), limit, offset. Returns: result

task_releaseC
Destructive

Release this executor's lease.

Inputs: token (required), task_id (required). Returns: result

task_updateB
DestructiveIdempotent

Update goals, constraints or policy with an explicit reason; immutable original request is preserved. A preflight budget uses policy.preflight_core_hours plus preflight_budget_authorization containing the user's confirmation. Estimates are not authorization. A budget change requires a new confirmation; wall_seconds is the scheduler's reservation per job, not predicted usage.

Inputs: op_id (required), reason (required), changes (required), task_id (required), +2 more. Returns: result

trajectory_correlateA
Read-onlyIdempotent

Link saved trajectory frames to exact thermo samples by Step and recorded Time, optionally selecting an executed phase (fix_id/style/group/marker/run_index/timestep). Preserve duplicate-step ambiguity; no interpolation or step*current-dt inference. Returns frame/cell/atom-type facts and actual matching rows for plots and structural previews.

Inputs: log (required), task_id (required), trajectory (required), phase. Returns: result

units_convertB
Read-onlyIdempotent

Convert dimensionally compatible scientific units, retaining the original expression (including ų).

Inputs: unit (required), value (required), target (required). Returns: result

validation_checkB
DestructiveIdempotent

Execute one frozen component on existing evidence. environment/numerics/outputs use run_id and collected data_root='runs/<run_id>'; a terminal failed job can supply complete usable components. Matching old runs can be rechecked under revised criteria without another simulation. A component with checker_id additionally uses analysis_operation_id from analysis_submit on that run's collected data; program version, qualified fixtures and raw data are verified. Static/analysis checks use project files. Returns a durable operation id; missing evidence affects this component only.

Inputs: op_id (required), plan_id (required), task_id (required), revision (required), +5 more. Returns: result

validation_compareC
Destructive

Compare candidate/final plans against the same inputs and actual receipts without new physics. Host agents may propose/check candidates concurrently, then select component criteria with a rationale in validation_plan(reuse_sources=[...]). Agreement is not measurement; preserve contrary evidence.

Inputs: task_id (required), plan_ids (required), snapshot_id (required). Returns: result

validation_controlC
DestructiveIdempotent

Suspend reuse of an invalid plan and its pending batches, retaining running evidence and costs. Corrected settings get a new plan with reuse_from for still valid components.

Inputs: op_id (required), reason (required), plan_id (required), task_id (required), +3 more. Returns: result

validation_matchC
Read-onlyIdempotent

Compare actual input dependencies, conditions, environment and current measurements. Returns direct/partial/needs_validation with exact missing components. Matching does not run a simulation.

Inputs: plan_id (required), task_id (required), snapshot_id (required). Returns: result

validation_planB
DestructiveIdempotent

Freeze task-specific criteria and native input dependencies. Components: id,scope (static/environment/numerics/outputs/analysis),files,condition_keys,checks; optional required=false for exploratory components, checker_id for a qualified native Python checker. Cover all inputs and conditions. Checks use role=required/advisory/candidate; only required evidence gates dependent work. Node monitor uses actual thermo phase selectors and action=warn (default) or explicit stop; no global physical tolerance. Changes retain rationale and parents. reuse_from or reuse_sources merges identical, valid components; changing a monitor preserves unrelated checks. Compare variants with validation_compare before selecting final criteria. variations: path,variable,min,max for seeds. Confirm policy.preflight_core_hours and preflight_budget_authorization; budget_core_hours defaults to that cap. runtime_id comes from environment_capture. Production requires static/environment/numerics/outputs; only missing checks need accounted preflight.

Inputs: entry (required), op_id (required), task_id (required), revision (required), +10 more. Returns: result

workflow_registerC
DestructiveIdempotent

Register an ordered native-script workflow. Each stage has id, kind (run/analysis), after IDs and spec. Run spec has run_id; analysis spec has script, inputs, outputs.

Inputs: op_id (required), stages (required), task_id (required), revision (required), +1 more. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    C2.8/5.0

    Scored across 75 tools

    Disambiguation2/5

    The tool set contains many overlapping read/get/check families: task_get, run_get, batch_get, check_get, operation_get, results_get, task_records, operation_read, response_read, file_read, and resource_file_read all retrieve state with subtle distinctions. Descriptions are long and do clarify boundaries, but an agent can easily select the wrong retrieval or status tool without careful cross-referencing.

    Naming Consistency4/5

    Most names follow a predictable snake_case noun_verb pattern with clear domain prefixes such as task_, run_, batch_, validation_, check_, resource_, and memory_. A few outliers like doctor, events, help, and response_read slightly break the pattern but remain readable.

    Tool Count1/5

    Seventy-five tools is far beyond a well-scoped set and imposes a very high selection burden. The domain is complex, but the surface is heavily fragmented into many narrowly split workflow, validation, retrieval, and background-control tools.

    Completeness4/5

    Coverage is broad: project lifecycle, resource registration, environment capture, validation planning, run/batch preparation and submission, analysis, artifact linking, memory, events, backup, and export/import are all present. Only minor gaps such as explicit deletion or simpler file/listing operations are apparent.

    Maintenance

    ActivityMaintained
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      B
      quality
      C
      maintenance
      Enables LLM agents to run verified DFT materials workflows (structure fetch, relaxation, band/DOS) with automated convergence gates and physics validation, ensuring every result is machine-verified with a complete evidence trail.
      5
      3
      MIT
    • A
      license
      B
      quality
      C
      maintenance
      Enables users to work on a SLURM HPC cluster over SSH through plain-language requests, covering file browsing and editing, batch job generation and submission, queue monitoring, and log reading to debug failed jobs. It runs with built-in safety guardrails such as fixed commands and configurable path policies for reads and writes.
      27
      MIT
    • A
      license
      B
      quality
      B
      maintenance
      Enables AI agents to plan, confirm, submit, observe, and audit compute jobs across local machines, SSH workstations (including WSL), and Slurm clusters through a unified stdio interface, with node probing, environment evidence, project sync, storage governance, and progress/ETA tracking.
      31
      1
      MIT