academic-army 0.4.0 → 0.5.0

package/README.md

@@ -38,10 +38,10 @@ Install package dependencies once:
 npm install
 ```
 
-Install MCP server dependencies from [`mcp-server/requirements.txt`](mcp-server/requirements.txt) if needed:
+Install the MCP server into the local virtual environment:
 
 ```bash
-python -m pip install -r ./mcp-server/requirements.txt
+.venv/bin/python -m pip install -e .
 ```
 
 ### 2. Configure AcademicArmy MCP
@@ -52,7 +52,7 @@ Create `.env` in the repository root:
 OPENAI_API_KEY=your_api_key_here
 ```
 
-For project pipeline runs, use the `academic_army_mcp_tools` server through [`agent-forge.yaml`](agent-forge.yaml). That config launches the server as `python -m mcp-server` with `PYTHONPATH=.` and `cwd=.` from the repository root, so the evolve/developing runners do not need a separate Codex MCP installation step.
+For project pipeline runs, use the `academic_army_mcp_tools` server through [`agent-forge.yaml`](agent-forge.yaml). That config launches the installed package with `.venv/bin/python -m academic_army_mcp_tools`, independent of each agent thread's working directory.
 
 When running AcademicArmy skills directly in Codex, use [`install_mcp.py`](install_mcp.py) to install the same MCP server into Codex so the skills can call `academic_army_mcp_tools.deepresearch` and `academic_army_mcp_tools.writing_master` outside the project pipeline:
 
@@ -70,16 +70,16 @@ Once you are satisfied with the paper blueprint, continue with the next planning
 
 ### 4. Run the development loop
 
-After the planning artifacts are ready, write the next high-level development objective into `workspace/goal.md`, then run:
+After the planning artifacts are ready, write the next high-level development objective into `workspace/plan/goal.md`, then run:
 
 ```bash
-$EDITOR workspace/goal.md
+$EDITOR workspace/plan/goal.md
 bash runs/develop.sh
 ```
 
-Run [`runs/develop.sh`](runs/develop.sh) to call the `developing-agent-forge` development pipeline, which reads the three planning artifacts plus the current `--goal-path` file and iteratively writes code under `workspace/codebase`. Each time you want the next new task, update `workspace/goal.md` before rerunning the wrapper.
+Run [`runs/develop.sh`](runs/develop.sh) to call the `developing-agent-forge` development pipeline, which reads the three planning artifacts plus the current `--goal-path` file and iteratively writes code under `workspace/codebase`. Each time you want the next new task, update `workspace/plan/goal.md` before rerunning the wrapper.
 
-The development loop stores project progress memory under `workspace/developing-memory/project-progress-memory` and code design memory under `workspace/developing-memory/code-design-memory`. If a new goal starts inheriting old context, edit or delete stale memory files in those directories before rerunning.
+The development loop stores project progress memory under `workspace/memory/project-progress` and code design memory under `workspace/memory/code-design`. If a new goal starts inheriting old context, edit or delete stale memory files in those directories before rerunning.
 
 See [`src/README.md`](src/README.md) for the local TypeScript entry points. The development loop implementation lives in the `developing-agent-forge` package.
 
@@ -103,11 +103,11 @@ npm run evolve-skill
 
 For the shared CLI and pipeline structure, see [`src/README.md`](src/README.md).
 
-For `developing` and `developing-skill`, update the file passed to `--goal-path` when you want to run a new task focus. The prepared wrappers use `workspace/goal.md`. The base development loop comes from `developing-agent-forge`; this repository provides the `developing-skill` trajectory optimizer locally.
+For `developing` and `developing-skill`, update the file passed to `--goal-path` when you want to run a new task focus. The prepared wrappers use `workspace/plan/goal.md`. The base development loop comes from `developing-agent-forge`; this repository provides the `developing-skill` trajectory optimizer locally.
 
 ### Call MCP Tools
 
-AcademicArmy includes a local stdio MCP implementation in the [`mcp-server`](mcp-server) directory. It exposes these tools:
+AcademicArmy includes an installable stdio MCP package in [`academic_army_mcp_tools`](academic_army_mcp_tools). It exposes these tools:
 
 - `deepresearch(prompt: str)`: runs the prompt with OpenAI Responses using `gpt-5.5`, high reasoning, web search, background mode, and source inclusion.
 - `writing_master(prompt: str)`: runs the prompt with OpenAI Responses using `gpt-5.5-pro`, high reasoning, web search, background mode, and source inclusion for high-end academic writing review.
@@ -121,16 +121,16 @@ Find the closest papers to this research idea, compare their methods, and return
 
 ## Project Structure
 
-| Path               | Purpose                                                                          |
-| ------------------ | -------------------------------------------------------------------------------- |
-| `agent-forge.yaml` | Agent and team wiring.                                                           |
-| `install_mcp.py`   | Installs the project MCP server into Codex for direct skill runs.                |
-| `mcp-server/`      | Local stdio MCP implementation that exposes `deepresearch` and `writing_master`. |
-| `skills/`          | Prepared AcademicArmy skills.                                                    |
-| `metaskills/`      | Matching metaskill design/evolution files.                                       |
-| `runs/`            | Convenience wrappers around TypeScript pipelines.                                |
-| `src/`             | TypeScript pipeline structure and implementation notes.                          |
-| `workspace/`       | Generated planning artifacts, codebase workspace, memory, and archives.          |
+| Path                       | Purpose                                                                                |
+| -------------------------- | -------------------------------------------------------------------------------------- |
+| `agent-forge.yaml`         | Agent and team wiring.                                                                 |
+| `install_mcp.py`           | Installs the project MCP server into Codex for direct skill runs.                      |
+| `academic_army_mcp_tools/` | Installable stdio MCP implementation that exposes `deepresearch` and `writing_master`. |
+| `skills/`                  | Prepared AcademicArmy skills.                                                          |
+| `metaskills/`              | Matching metaskill design/evolution files.                                             |
+| `runs/`                    | Convenience wrappers around TypeScript pipelines.                                      |
+| `src/`                     | TypeScript pipeline structure and implementation notes.                                |
+| `workspace/`               | Generated planning artifacts, codebase workspace, memory, and archives.                |
 
 Agent and team wiring lives in [`agent-forge.yaml`](agent-forge.yaml). The local TypeScript agents are implemented under [`src/evolve-skill/agents`](src/evolve-skill/agents); developing agents come from `developing-agent-forge`.
 
@@ -141,7 +141,7 @@ Prepared AcademicArmy skills live under [`skills/`](skills/), and their matching
 | File or variable          | Required for           | Notes                                                                                                                                                                                          |
 | ------------------------- | ---------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | `.env` / `OPENAI_API_KEY` | AcademicArmy MCP       | Read by the MCP server and by `install_mcp.py`.                                                                                                                                                |
-| `agent-forge.yaml`        | Project pipelines      | Launches `academic_army_mcp_tools` as `python -m mcp-server` with `PYTHONPATH=.` and `cwd=.`.                                                                                                  |
+| `agent-forge.yaml`        | Project pipelines      | Launches installed `.venv` package `academic_army_mcp_tools`; run `.venv/bin/python -m pip install -e .` after changing MCP package code.                                                      |
 | `secret.yaml`             | Prepared shell scripts | Local ignored config overlay used by the prepared wrappers. It may contain passwords, API keys, runtime credentials, or other private values that must not be committed or uploaded to GitHub. |
 
 To override or add environment variables directly when installing MCP into Codex, repeat `-e/--env NAME=VALUE`:
@@ -150,7 +150,7 @@ To override or add environment variables directly when installing MCP into Codex
 python install_mcp.py -e OPENAI_API_KEY=your_api_key_here
 ```
 
-Running the installer refreshes the Codex `academic_army_mcp_tools` entry, registers the current Python executable with `-m mcp-server`, sets the repository root as the MCP working directory, reads `.env`, and forwards those values to the MCP server.
+Running the installer refreshes the Codex `academic_army_mcp_tools` entry, registers the current Python executable with `-m academic_army_mcp_tools`, sets the repository root as the MCP working directory, reads `.env`, and forwards those values to the MCP server.
 
 ## Troubleshooting
 

package/README.zh-CN.md

@@ -38,10 +38,10 @@ AcademicArmy 的主体核心可以概括为一句话：按图施工。
 npm install
 ```
 
-如有需要，从 [`mcp-server/requirements.txt`](mcp-server/requirements.txt) 安装 MCP server 依赖：
+把 MCP server 安装到本地虚拟环境：
 
 ```bash
-python -m pip install -r ./mcp-server/requirements.txt
+.venv/bin/python -m pip install -e .
 ```
 
 ### 2. 配置 AcademicArmy MCP
@@ -52,7 +52,7 @@ python -m pip install -r ./mcp-server/requirements.txt
 OPENAI_API_KEY=your_api_key_here
 ```
 
-运行项目 pipeline 时，通过 [`agent-forge.yaml`](agent-forge.yaml) 使用 `academic_army_mcp_tools`。该配置会在仓库根目录以 `PYTHONPATH=.` 和 `cwd=.` 运行 `python -m mcp-server`，因此 evolve/developing runner 不需要额外执行 Codex MCP 安装步骤。
+运行项目 pipeline 时，通过 [`agent-forge.yaml`](agent-forge.yaml) 使用 `academic_army_mcp_tools`。该配置会用 `.venv/bin/python -m academic_army_mcp_tools` 启动已安装的包，不受各个 agent thread 工作目录影响。
 
 如果直接在 Codex 中运行 AcademicArmy skills，需要用 [`install_mcp.py`](install_mcp.py) 把同一个 MCP server 安装到 Codex 里，这样 skill 才能在项目 pipeline 之外调用 `academic_army_mcp_tools.deepresearch` 和 `academic_army_mcp_tools.writing_master`：
 
@@ -70,16 +70,16 @@ python install_mcp.py
 
 ### 4. 运行开发循环
 
-三份规划产物准备好后，先把下一轮高层开发目标写进 `workspace/goal.md`，然后运行：
+三份规划产物准备好后，先把下一轮高层开发目标写进 `workspace/plan/goal.md`，然后运行：
 
 ```bash
-$EDITOR workspace/goal.md
+$EDITOR workspace/plan/goal.md
 bash runs/develop.sh
 ```
 
-运行 [`runs/develop.sh`](runs/develop.sh) 来调用 `developing-agent-forge` 提供的 `developing` pipeline，读取三份规划产物和当前 `--goal-path` 文件，并在 `workspace/codebase` 下迭代写代码。每次想执行下一个新任务时，先更新 `workspace/goal.md`，再重新运行 wrapper。
+运行 [`runs/develop.sh`](runs/develop.sh) 来调用 `developing-agent-forge` 提供的 `developing` pipeline，读取三份规划产物和当前 `--goal-path` 文件，并在 `workspace/codebase` 下迭代写代码。每次想执行下一个新任务时，先更新 `workspace/plan/goal.md`，再重新运行 wrapper。
 
-开发循环会把项目进度记忆放在 `workspace/developing-memory/project-progress-memory`，把代码设计记忆放在 `workspace/developing-memory/code-design-memory`。如果新 goal 开始继承旧上下文，可以在重新运行前编辑或删除这些目录里的过期 memory 文件。
+开发循环会把项目进度记忆放在 `workspace/memory/project-progress`，把代码设计记忆放在 `workspace/memory/code-design`。如果新 goal 开始继承旧上下文，可以在重新运行前编辑或删除这些目录里的过期 memory 文件。
 
 本仓库的 TypeScript 入口和目录结构见 [`src/README.zh-CN.md`](src/README.zh-CN.md)。开发循环实现来自 `developing-agent-forge` 包。
 
@@ -103,11 +103,11 @@ npm run evolve-skill
 
 TypeScript pipeline 的目录结构和实现说明见 [`src/README.zh-CN.md`](src/README.zh-CN.md)。
 
-对于 `developing` 和 `developing-skill`，每次想切换到新的任务重点时，更新传给 `--goal-path` 的文件即可。预设 wrappers 使用的是 `workspace/goal.md`。基础开发循环来自 `developing-agent-forge`；本仓库在本地提供 `developing-skill` trajectory optimizer。
+对于 `developing` 和 `developing-skill`，每次想切换到新的任务重点时，更新传给 `--goal-path` 的文件即可。预设 wrappers 使用的是 `workspace/plan/goal.md`。基础开发循环来自 `developing-agent-forge`；本仓库在本地提供 `developing-skill` trajectory optimizer。
 
 ### 调用 MCP 工具
 
-AcademicArmy 在 [`mcp-server`](mcp-server) 目录下提供了本地 stdio MCP 实现。它暴露这些工具：
+AcademicArmy 在 [`academic_army_mcp_tools`](academic_army_mcp_tools) 包中提供了可安装的 stdio MCP 实现。它暴露这些工具：
 
 - `deepresearch(prompt: str)`：把 prompt 交给 OpenAI Responses，以 `gpt-5.5`、high reasoning、web search、background mode 和 source inclusion 的固定配置运行。
 - `writing_master(prompt: str)`：把 prompt 交给 OpenAI Responses，以 `gpt-5.5-pro`、high reasoning、web search、background mode 和 source inclusion 的固定配置运行，用于高阶学术写作咨询。
@@ -121,16 +121,16 @@ Find the closest papers to this research idea, compare their methods, and return
 
 ## 项目结构
 
-| 路径               | 用途                                                           |
-| ------------------ | -------------------------------------------------------------- |
-| `agent-forge.yaml` | Agent 和团队 wiring。                                          |
-| `install_mcp.py`   | 把项目 MCP server 安装到 Codex，供直接运行 skill 时使用。      |
-| `mcp-server/`      | 本地 stdio MCP 实现，暴露 `deepresearch` 和 `writing_master`。 |
-| `skills/`          | 已准备的 AcademicArmy skills。                                 |
-| `metaskills/`      | 对应的 metaskill 设计与 evolution 文件。                       |
-| `runs/`            | TypeScript pipelines 的便捷 wrappers。                         |
-| `src/`             | TypeScript pipeline 的目录结构和实现说明。                     |
-| `workspace/`       | 生成的规划产物、代码库工作区、记忆和归档。                     |
+| 路径                       | 用途                                                               |
+| -------------------------- | ------------------------------------------------------------------ |
+| `agent-forge.yaml`         | Agent 和团队 wiring。                                              |
+| `install_mcp.py`           | 把项目 MCP server 安装到 Codex，供直接运行 skill 时使用。          |
+| `academic_army_mcp_tools/` | 可安装的 stdio MCP 实现，暴露 `deepresearch` 和 `writing_master`。 |
+| `skills/`                  | 已准备的 AcademicArmy skills。                                     |
+| `metaskills/`              | 对应的 metaskill 设计与 evolution 文件。                           |
+| `runs/`                    | TypeScript pipelines 的便捷 wrappers。                             |
+| `src/`                     | TypeScript pipeline 的目录结构和实现说明。                         |
+| `workspace/`               | 生成的规划产物、代码库工作区、记忆和归档。                         |
 
 Agent 和团队 wiring 位于 [`agent-forge.yaml`](agent-forge.yaml)。本仓库内的 TypeScript agents 实现于 [`src/evolve-skill/agents`](src/evolve-skill/agents)；developing agents 来自 `developing-agent-forge`。
 
@@ -141,7 +141,7 @@ Agent 和团队 wiring 位于 [`agent-forge.yaml`](agent-forge.yaml)。本仓库
 | 文件或变量                | 用于               | 说明                                                                                                                     |
 | ------------------------- | ------------------ | ------------------------------------------------------------------------------------------------------------------------ |
 | `.env` / `OPENAI_API_KEY` | AcademicArmy MCP   | MCP server 和 `install_mcp.py` 会读取。                                                                                  |
-| `agent-forge.yaml`        | 项目 pipelines     | 以 `PYTHONPATH=.` 和 `cwd=.` 运行 `python -m mcp-server`。                                                               |
+| `agent-forge.yaml`        | 项目 pipelines     | 启动 `.venv` 中已安装的 `academic_army_mcp_tools` 包；修改 MCP 包代码后运行 `.venv/bin/python -m pip install -e .`。     |
 | `secret.yaml`             | 预设 shell scripts | 预设 wrappers 使用的本地忽略 config overlay。它可以包含密码、API key、runtime 凭据等不能提交或上传到 GitHub 的隐私内容。 |
 
 如果需要覆盖或补充环境变量，可以重复使用 `-e/--env NAME=VALUE`：
@@ -150,7 +150,7 @@ Agent 和团队 wiring 位于 [`agent-forge.yaml`](agent-forge.yaml)。本仓库
 python install_mcp.py -e OPENAI_API_KEY=your_api_key_here
 ```
 
-运行安装脚本时，会刷新 Codex 中的 `academic_army_mcp_tools` 配置项，注册当前 Python 可执行文件和 `-m mcp-server`，把仓库根目录设置为 MCP 工作目录，读取 `.env`，并把这些环境变量传给 MCP server。
+运行安装脚本时，会刷新 Codex 中的 `academic_army_mcp_tools` 配置项，注册当前 Python 可执行文件和 `-m academic_army_mcp_tools`，把仓库根目录设置为 MCP 工作目录，读取 `.env`，并把这些环境变量传给 MCP server。
 
 ## 常见问题
 

package/academic_army_mcp_tools/__init__.py

@@ -0,0 +1 @@
+"""AcademicArmy MCP server package."""

package/{mcp-server → academic_army_mcp_tools}/__main__.py

@@ -8,16 +8,7 @@ from .deepresearch import register_deepresearch
 from .writing_master import register_writing_master
 
 
-mcp = FastMCP(
-    "academic-army",
-    instructions=(
-        "AcademicArmy MCP server. It exposes project-level tools for research, "
-        "blueprint orchestration, and future AcademicArmy workflow functions."
-    ),
-)
-
-
-if __name__ == "__main__":
+def main() -> None:
     load_dotenv(".env")
 
     parser = argparse.ArgumentParser()
@@ -30,6 +21,17 @@ if __name__ == "__main__":
             parser.error("-e/--env must be NAME=VALUE")
         os.environ[name] = value
 
+    mcp = FastMCP(
+        "academic-army",
+        instructions=(
+            "AcademicArmy MCP server. It exposes project-level tools for research, "
+            "blueprint orchestration, and future AcademicArmy workflow functions."
+        ),
+    )
     register_deepresearch(mcp)
     register_writing_master(mcp)
     mcp.run(transport="stdio")
+
+
+if __name__ == "__main__":
+    main()

package/agent-forge.yaml

@@ -8,10 +8,8 @@ runtimes:
             command: python
             args:
               - -m
-              - mcp-server
+              - academic_army_mcp_tools
             cwd: .
-            env:
-              PYTHONPATH: .
             tool_timeout_sec: 3600
   deepseek-v4-pro:
     kind: claude
@@ -31,10 +29,8 @@ runtimes:
           command: python
           args:
             - -m
-            - mcp-server
+            - academic_army_mcp_tools
           cwd: .
-          env:
-            PYTHONPATH: .
           timeout: 3600000
   deepseek-v4-flash:
     kind: claude
@@ -97,7 +93,7 @@ threads:
   memory:
     runtime: deepseek-v4-flash
     options:
-      cwd: &memoryWorkingDir workspace/developing-memory
+      cwd: &memoryWorkingDir workspace/memory
       permissionMode: dontAsk
       allowedTools:
         - Read

package/dist/developing-skill/agents/prompts.d.ts

@@ -0,0 +1,2 @@
+export declare function quoteBlock(content: string): string;
+//# sourceMappingURL=prompts.d.ts.map

package/dist/developing-skill/agents/prompts.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"prompts.d.ts","sourceRoot":"","sources":["../../../src/developing-skill/agents/prompts.ts"],"names":[],"mappings":"AAAA,wBAAgB,UAAU,CAAC,OAAO,EAAE,MAAM,GAAG,MAAM,CAKlD"}

package/dist/developing-skill/agents/prompts.js

@@ -0,0 +1,7 @@
+export function quoteBlock(content) {
+    return content
+        .split(/\r?\n/)
+        .map((line) => `> ${line}`)
+        .join("\n");
+}
+//# sourceMappingURL=prompts.js.map

package/dist/developing-skill/agents/prompts.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"prompts.js","sourceRoot":"","sources":["../../../src/developing-skill/agents/prompts.ts"],"names":[],"mappings":"AAAA,MAAM,UAAU,UAAU,CAAC,OAAe;IACxC,OAAO,OAAO;SACX,KAAK,CAAC,OAAO,CAAC;SACd,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,KAAK,IAAI,EAAE,CAAC;SAC1B,IAAI,CAAC,IAAI,CAAC,CAAC;AAChB,CAAC"}

package/dist/developing-skill/agents/trajectory-optimizer.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"trajectory-optimizer.d.ts","sourceRoot":"","sources":["../../../src/developing-skill/agents/trajectory-optimizer.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AACjD,OAAO,KAAK,EAAE,wBAAwB,EAAE,MAAM,+BAA+B,CAAC;AAE9E,MAAM,MAAM,4BAA4B,GAAG,wBAAwB,GAAG;IACpE,oBAAoB,EAAE,MAAM,CAAC;IAC7B,SAAS,EAAE,MAAM,CAAC;IAClB,gBAAgB,EAAE,MAAM,CAAC;IACzB,SAAS,EAAE,MAAM,CAAC;CACnB,CAAC;AAEF,qBAAa,wBAAyB,SAAQ,KAAK,CAAC,4BAA4B,CAAC;IAC/E,SAAS,CAAC,WAAW,CAAC,SAAS,EAAE,QAAQ,CAAC,4BAA4B,CAAC,GAAG,MAAM;CA0BjF"}
+{"version":3,"file":"trajectory-optimizer.d.ts","sourceRoot":"","sources":["../../../src/developing-skill/agents/trajectory-optimizer.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AACjD,OAAO,KAAK,EAAE,wBAAwB,EAAE,MAAM,+BAA+B,CAAC;AAI9E,MAAM,MAAM,4BAA4B,GAAG,wBAAwB,GAAG;IACpE,oBAAoB,EAAE,MAAM,CAAC;IAC7B,SAAS,EAAE,MAAM,CAAC;IAClB,gBAAgB,EAAE,MAAM,CAAC;IACzB,SAAS,EAAE,MAAM,CAAC;CACnB,CAAC;AAEF,qBAAa,wBAAyB,SAAQ,KAAK,CAAC,4BAA4B,CAAC;IAC/E,SAAS,CAAC,WAAW,CAAC,SAAS,EAAE,QAAQ,CAAC,4BAA4B,CAAC,GAAG,MAAM;CA+BjF"}

package/dist/developing-skill/agents/trajectory-optimizer.js

@@ -1,29 +1,35 @@
 import { Agent } from "coding-agent-forge/agent";
+import { quoteBlock } from "./prompts.js";
 export class TrajectoryOptimizerAgent extends Agent {
     buildPrompt(variables) {
         return `
-Revise the skill at ${variables.codingStyleSkillPath} so it produces better development trajectories.
+Evaluate and revise the skill so it produces better development trajectories.
 
-The metaskill below contains the design goals and tips of this skill:
+Skill path: ${variables.codingStyleSkillPath}
+Project root: ${variables.targetPath}/.
 
-${variables.metaskill}
+Current goal:
+${quoteBlock(variables.goal)}
 
-The sections below describe the task this skill just executed and what happened during that round.
+Task brief:
+${quoteBlock(variables.taskBrief)}
 
-Target repository at ${variables.targetPath}/.
+Round summary:
+${quoteBlock(variables.taskRoundSummary)}
 
-Goal:
-${variables.goal}
+Metaskill (design goals and tips for this skill):
+${quoteBlock(variables.metaskill)}
 
-Task Brief:
-${variables.taskBrief}
+Consider these design goals and tips when revising.
 
-Reality-aware task round summary:
-${variables.taskRoundSummary}
+Evaluate whether the skill produced a good modification trajectory, then revise the skill directly.
 
-Evaluate whether the skill produced a good modification trajectory, then edit the skill directly. Focus on missing, misleading, or redundant guidance that affected task selection, coding, or review.
-
-Output a concise optimizer report with the main skill changes.
+Useful angles include:
+- missing guidance
+- misleading guidance
+- redundant guidance
+- effects on task selection, coding, or review
+- other relevant trajectory issues...
 `;
     }
 }

package/dist/developing-skill/agents/trajectory-optimizer.js.map

@@ -1 +1 @@
-{"version":3,"file":"trajectory-optimizer.js","sourceRoot":"","sources":["../../../src/developing-skill/agents/trajectory-optimizer.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAUjD,MAAM,OAAO,wBAAyB,SAAQ,KAAmC;IACrE,WAAW,CAAC,SAAiD;QACrE,OAAO;sBACW,SAAS,CAAC,oBAAoB;;;;EAIlD,SAAS,CAAC,SAAS;;;;uBAIE,SAAS,CAAC,UAAU;;;EAGzC,SAAS,CAAC,IAAI;;;EAGd,SAAS,CAAC,SAAS;;;EAGnB,SAAS,CAAC,gBAAgB;;;;;CAK3B,CAAC;IACA,CAAC;CACF"}
+{"version":3,"file":"trajectory-optimizer.js","sourceRoot":"","sources":["../../../src/developing-skill/agents/trajectory-optimizer.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAGjD,OAAO,EAAE,UAAU,EAAE,MAAM,cAAc,CAAC;AAS1C,MAAM,OAAO,wBAAyB,SAAQ,KAAmC;IACrE,WAAW,CAAC,SAAiD;QACrE,OAAO;;;cAGG,SAAS,CAAC,oBAAoB;gBAC5B,SAAS,CAAC,UAAU;;;EAGlC,UAAU,CAAC,SAAS,CAAC,IAAI,CAAC;;;EAG1B,UAAU,CAAC,SAAS,CAAC,SAAS,CAAC;;;EAG/B,UAAU,CAAC,SAAS,CAAC,gBAAgB,CAAC;;;EAGtC,UAAU,CAAC,SAAS,CAAC,SAAS,CAAC;;;;;;;;;;;;CAYhC,CAAC;IACA,CAAC;CACF"}

package/dist/developing-skill/pipeline.d.ts

@@ -11,7 +11,8 @@ export declare const developingSkillArgsOptions: {
     };
     readonly "goal-path": {
         readonly type: "string";
-        readonly description: "Goal document path";
+        readonly multiple: true;
+        readonly description: "Goal document path; repeat to concatenate multiple goals in order";
     };
     readonly "achive-dir": {
         readonly type: "string";
@@ -63,7 +64,8 @@ export declare const developingSkillPipeline: import("coding-agent-forge").Pipel
     };
     readonly "goal-path": {
         readonly type: "string";
-        readonly description: "Goal document path";
+        readonly multiple: true;
+        readonly description: "Goal document path; repeat to concatenate multiple goals in order";
     };
     readonly "achive-dir": {
         readonly type: "string";

package/dist/developing-skill/pipeline.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"pipeline.d.ts","sourceRoot":"","sources":["../../src/developing-skill/pipeline.ts"],"names":[],"mappings":"AAAA,OAAO,EACL,SAAS,EAGT,KAAK,eAAe,EAErB,MAAM,oBAAoB,CAAC;AAC5B,OAAO,EAGL,KAAK,kCAAkC,EAExC,MAAM,wBAAwB,CAAC;AAGhC,OAAO,KAAK,EAAE,4BAA4B,EAAE,MAAM,mBAAmB,CAAC;AAEtE,MAAM,MAAM,6BAA6B,GAAG,kCAAkC,GAAG;IAC/E,sBAAsB,EAAE,4BAA4B,CAAC;CACtD,CAAC;AAEF,eAAO,MAAM,0BAA0B;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAUC,CAAC;AAEzC,MAAM,MAAM,sBAAsB,GAAG,eAAe,CAAC,OAAO,0BAA0B,CAAC,CAAC;AAExF,wBAAsB,eAAe,CACnC,IAAI,EAAE,SAAS,CAAC,6BAA6B,CAAC,EAC9C,OAAO,EAAE,sBAAsB,GAC9B,OAAO,CAAC,IAAI,CAAC,CAqCf;AAED,eAAO,MAAM,uBAAuB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;iCAQlC,CAAC"}
+{"version":3,"file":"pipeline.d.ts","sourceRoot":"","sources":["../../src/developing-skill/pipeline.ts"],"names":[],"mappings":"AAAA,OAAO,EACL,SAAS,EAGT,KAAK,eAAe,EAErB,MAAM,oBAAoB,CAAC;AAC5B,OAAO,EAGL,KAAK,kCAAkC,EAExC,MAAM,wBAAwB,CAAC;AAGhC,OAAO,KAAK,EAAE,4BAA4B,EAAE,MAAM,mBAAmB,CAAC;AAEtE,MAAM,MAAM,6BAA6B,GAAG,kCAAkC,GAAG;IAC/E,sBAAsB,EAAE,4BAA4B,CAAC;CACtD,CAAC;AAEF,eAAO,MAAM,0BAA0B;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAUC,CAAC;AAEzC,MAAM,MAAM,sBAAsB,GAAG,eAAe,CAAC,OAAO,0BAA0B,CAAC,CAAC;AAExF,wBAAsB,eAAe,CACnC,IAAI,EAAE,SAAS,CAAC,6BAA6B,CAAC,EAC9C,OAAO,EAAE,sBAAsB,GAC9B,OAAO,CAAC,IAAI,CAAC,CAqCf;AAED,eAAO,MAAM,uBAAuB;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;iCAQlC,CAAC"}

package/dist/evolve-skill/agents/evaluator.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"evaluator.d.ts","sourceRoot":"","sources":["../../../src/evolve-skill/agents/evaluator.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAEjD,MAAM,MAAM,uBAAuB,GAAG;IACpC,YAAY,EAAE,MAAM,CAAC;IACrB,SAAS,EAAE,MAAM,CAAC;IAClB,gBAAgB,EAAE,MAAM,CAAC;CAC1B,CAAC;AAEF,qBAAa,mBAAoB,SAAQ,KAAK,CAAC,uBAAuB,CAAC;IACrE,SAAS,CAAC,WAAW,CAAC,SAAS,EAAE,QAAQ,CAAC,uBAAuB,CAAC,GAAG,MAAM;CAgB5E"}
+{"version":3,"file":"evaluator.d.ts","sourceRoot":"","sources":["../../../src/evolve-skill/agents/evaluator.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAIjD,MAAM,MAAM,uBAAuB,GAAG;IACpC,YAAY,EAAE,MAAM,CAAC;IACrB,SAAS,EAAE,MAAM,CAAC;IAClB,gBAAgB,EAAE,MAAM,CAAC;CAC1B,CAAC;AAEF,qBAAa,mBAAoB,SAAQ,KAAK,CAAC,uBAAuB,CAAC;IACrE,SAAS,CAAC,WAAW,CAAC,SAAS,EAAE,QAAQ,CAAC,uBAAuB,CAAC,GAAG,MAAM;CAuB5E"}

package/dist/evolve-skill/agents/evaluator.js

@@ -1,19 +1,27 @@
 import { Agent } from "coding-agent-forge/agent";
+import { quoteBlock } from "./prompts.js";
 export class SkillEvaluatorAgent extends Agent {
     buildPrompt(variables) {
         return `
-Evaluate the artifact at ${variables.artifactPath}. It was produced by a skill.
+Evaluate the artifact produced by this skill.
 
-The artifacts were created based on the following task descriptions:
+Artifact path: ${variables.artifactPath}
 
-${variables.taskDescriptions}
+Task descriptions:
+${quoteBlock(variables.taskDescriptions)}
 
-The metaskill below contains the design goals and tips of this skill:
+Metaskill (design goals and tips for this skill):
+${quoteBlock(variables.metaskill)}
 
-${variables.metaskill}
+Review the artifact against the metaskill.
 
-Based on these goals and tips, are there any problems in the artifact produced by this skill? Are there any redundant parts?
-Carefully inspect both the language and the content, and use that analysis to explain how this skill can be optimized.
+Useful angles include:
+- problems in content or language
+- missing or misleading guidance
+- redundant parts
+- other relevant issues...
+
+Return a concise report that explains how this skill can be optimized.
 `;
     }
 }

package/dist/evolve-skill/agents/evaluator.js.map

@@ -1 +1 @@
-{"version":3,"file":"evaluator.js","sourceRoot":"","sources":["../../../src/evolve-skill/agents/evaluator.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAQjD,MAAM,OAAO,mBAAoB,SAAQ,KAA8B;IAC3D,WAAW,CAAC,SAA4C;QAChE,OAAO;2BACgB,SAAS,CAAC,YAAY;;;;EAI/C,SAAS,CAAC,gBAAgB;;;;EAI1B,SAAS,CAAC,SAAS;;;;CAIpB,CAAC;IACA,CAAC;CACF"}
+{"version":3,"file":"evaluator.js","sourceRoot":"","sources":["../../../src/evolve-skill/agents/evaluator.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAEjD,OAAO,EAAE,UAAU,EAAE,MAAM,cAAc,CAAC;AAQ1C,MAAM,OAAO,mBAAoB,SAAQ,KAA8B;IAC3D,WAAW,CAAC,SAA4C;QAChE,OAAO;;;iBAGM,SAAS,CAAC,YAAY;;;EAGrC,UAAU,CAAC,SAAS,CAAC,gBAAgB,CAAC;;;EAGtC,UAAU,CAAC,SAAS,CAAC,SAAS,CAAC;;;;;;;;;;;CAWhC,CAAC;IACA,CAAC;CACF"}

package/dist/evolve-skill/agents/modifier.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"modifier.d.ts","sourceRoot":"","sources":["../../../src/evolve-skill/agents/modifier.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAEjD,MAAM,MAAM,sBAAsB,GAAG;IACnC,SAAS,EAAE,MAAM,CAAC;IAClB,SAAS,EAAE,MAAM,CAAC;IAClB,MAAM,EAAE,MAAM,CAAC;CAChB,CAAC;AAEF,qBAAa,kBAAmB,SAAQ,KAAK,CAAC,sBAAsB,CAAC;IACnE,SAAS,CAAC,WAAW,CAAC,SAAS,EAAE,QAAQ,CAAC,sBAAsB,CAAC,GAAG,MAAM;CAe3E"}
+{"version":3,"file":"modifier.d.ts","sourceRoot":"","sources":["../../../src/evolve-skill/agents/modifier.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAIjD,MAAM,MAAM,sBAAsB,GAAG;IACnC,SAAS,EAAE,MAAM,CAAC;IAClB,SAAS,EAAE,MAAM,CAAC;IAClB,MAAM,EAAE,MAAM,CAAC;CAChB,CAAC;AAEF,qBAAa,kBAAmB,SAAQ,KAAK,CAAC,sBAAsB,CAAC;IACnE,SAAS,CAAC,WAAW,CAAC,SAAS,EAAE,QAAQ,CAAC,sBAAsB,CAAC,GAAG,MAAM;CAiB3E"}

package/dist/evolve-skill/agents/modifier.js

@@ -1,18 +1,21 @@
 import { Agent } from "coding-agent-forge/agent";
+import { quoteBlock } from "./prompts.js";
 export class SkillModifierAgent extends Agent {
     buildPrompt(variables) {
         return `
-Revise the skill at ${variables.skillPath} using the feedback below. The feedback is based on an artifact produced by this skill.
+Revise the skill using the evaluation feedback.
 
-The metaskill below contains the design goals and tips of this skill:
+Skill path: ${variables.skillPath}
 
-${variables.metaskill}
+Metaskill (design goals and tips for this skill):
+${quoteBlock(variables.metaskill)}
 
 Consider these design goals and tips when revising.
 
-Feedback:
+Evaluation feedback:
+${quoteBlock(variables.review)}
 
-${variables.review}
+Edit the skill directly.
 `;
     }
 }

package/dist/evolve-skill/agents/modifier.js.map

@@ -1 +1 @@
-{"version":3,"file":"modifier.js","sourceRoot":"","sources":["../../../src/evolve-skill/agents/modifier.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAQjD,MAAM,OAAO,kBAAmB,SAAQ,KAA6B;IACzD,WAAW,CAAC,SAA2C;QAC/D,OAAO;sBACW,SAAS,CAAC,SAAS;;;;EAIvC,SAAS,CAAC,SAAS;;;;;;EAMnB,SAAS,CAAC,MAAM;CACjB,CAAC;IACA,CAAC;CACF"}
+{"version":3,"file":"modifier.js","sourceRoot":"","sources":["../../../src/evolve-skill/agents/modifier.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAEjD,OAAO,EAAE,UAAU,EAAE,MAAM,cAAc,CAAC;AAQ1C,MAAM,OAAO,kBAAmB,SAAQ,KAA6B;IACzD,WAAW,CAAC,SAA2C;QAC/D,OAAO;;;cAGG,SAAS,CAAC,SAAS;;;EAG/B,UAAU,CAAC,SAAS,CAAC,SAAS,CAAC;;;;;EAK/B,UAAU,CAAC,SAAS,CAAC,MAAM,CAAC;;;CAG7B,CAAC;IACA,CAAC;CACF"}

package/dist/evolve-skill/agents/prompts.d.ts

@@ -0,0 +1,2 @@
+export declare function quoteBlock(content: string): string;
+//# sourceMappingURL=prompts.d.ts.map

package/dist/evolve-skill/agents/prompts.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"prompts.d.ts","sourceRoot":"","sources":["../../../src/evolve-skill/agents/prompts.ts"],"names":[],"mappings":"AAAA,wBAAgB,UAAU,CAAC,OAAO,EAAE,MAAM,GAAG,MAAM,CAKlD"}

package/dist/evolve-skill/agents/prompts.js

@@ -0,0 +1,7 @@
+export function quoteBlock(content) {
+    return content
+        .split(/\r?\n/)
+        .map((line) => `> ${line}`)
+        .join("\n");
+}
+//# sourceMappingURL=prompts.js.map

package/dist/evolve-skill/agents/prompts.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"prompts.js","sourceRoot":"","sources":["../../../src/evolve-skill/agents/prompts.ts"],"names":[],"mappings":"AAAA,MAAM,UAAU,UAAU,CAAC,OAAe;IACxC,OAAO,OAAO;SACX,KAAK,CAAC,OAAO,CAAC;SACd,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,KAAK,IAAI,EAAE,CAAC;SAC1B,IAAI,CAAC,IAAI,CAAC,CAAC;AAChB,CAAC"}

package/dist/evolve-skill/agents/runner.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"runner.d.ts","sourceRoot":"","sources":["../../../src/evolve-skill/agents/runner.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAGjD,MAAM,MAAM,oBAAoB,GAAG;IACjC,SAAS,EAAE,MAAM,CAAC;IAClB,YAAY,EAAE,MAAM,CAAC;IACrB,QAAQ,EAAE,MAAM,CAAC;CAClB,CAAC;AAEF,qBAAa,gBAAiB,SAAQ,KAAK,CAAC,oBAAoB,CAAC;IAC/D,SAAS,CAAC,WAAW,CAAC,SAAS,EAAE,QAAQ,CAAC,oBAAoB,CAAC,GAAG,MAAM;CAQzE"}
+{"version":3,"file":"runner.d.ts","sourceRoot":"","sources":["../../../src/evolve-skill/agents/runner.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AAKjD,MAAM,MAAM,oBAAoB,GAAG;IACjC,SAAS,EAAE,MAAM,CAAC;IAClB,YAAY,EAAE,MAAM,CAAC;IACrB,QAAQ,EAAE,MAAM,CAAC;CAClB,CAAC;AAEF,qBAAa,gBAAiB,SAAQ,KAAK,CAAC,oBAAoB,CAAC;IAC/D,SAAS,CAAC,WAAW,CAAC,SAAS,EAAE,QAAQ,CAAC,oBAAoB,CAAC,GAAG,MAAM;CAczE"}

package/dist/evolve-skill/agents/runner.js

@@ -1,12 +1,19 @@
 import { Agent } from "coding-agent-forge/agent";
 import { readFileSync } from "node:fs";
+import { quoteBlock } from "./prompts.js";
 export class SkillRunnerAgent extends Agent {
     buildPrompt(variables) {
         const task = readFileSync(variables.taskPath, "utf8");
         return `
-Use the skill at ${variables.skillPath} to complete the task below. Save all relevant output files in ${variables.artifactPath}.
+Use the skill to complete the task.
 
-${task}
+Skill path: ${variables.skillPath}
+Artifact path: ${variables.artifactPath}
+
+Task:
+${quoteBlock(task)}
+
+Save all relevant output files in the artifact path.
 `;
     }
 }

package/dist/evolve-skill/agents/runner.js.map

@@ -1 +1 @@
-{"version":3,"file":"runner.js","sourceRoot":"","sources":["../../../src/evolve-skill/agents/runner.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AACjD,OAAO,EAAE,YAAY,EAAE,MAAM,SAAS,CAAC;AAQvC,MAAM,OAAO,gBAAiB,SAAQ,KAA2B;IACrD,WAAW,CAAC,SAAyC;QAC7D,MAAM,IAAI,GAAG,YAAY,CAAC,SAAS,CAAC,QAAQ,EAAE,MAAM,CAAC,CAAC;QACtD,OAAO;mBACQ,SAAS,CAAC,SAAS,kEAAkE,SAAS,CAAC,YAAY;;EAE5H,IAAI;CACL,CAAC;IACA,CAAC;CACF"}
+{"version":3,"file":"runner.js","sourceRoot":"","sources":["../../../src/evolve-skill/agents/runner.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,MAAM,0BAA0B,CAAC;AACjD,OAAO,EAAE,YAAY,EAAE,MAAM,SAAS,CAAC;AAEvC,OAAO,EAAE,UAAU,EAAE,MAAM,cAAc,CAAC;AAQ1C,MAAM,OAAO,gBAAiB,SAAQ,KAA2B;IACrD,WAAW,CAAC,SAAyC;QAC7D,MAAM,IAAI,GAAG,YAAY,CAAC,SAAS,CAAC,QAAQ,EAAE,MAAM,CAAC,CAAC;QACtD,OAAO;;;cAGG,SAAS,CAAC,SAAS;iBAChB,SAAS,CAAC,YAAY;;;EAGrC,UAAU,CAAC,IAAI,CAAC;;;CAGjB,CAAC;IACA,CAAC;CACF"}

package/install_mcp.py

@@ -44,8 +44,6 @@ for item in args.env:
     name, value = item.split("=", 1)
     env_items[name] = value
 
-env_items.setdefault("PYTHONPATH", str(repo))
-
 codex = shutil.which("codex")
 if not codex:
     raise SystemExit("Could not find the codex command line tool in PATH.")
@@ -59,7 +57,7 @@ subprocess.run(
 command = [codex, "mcp", "add"]
 for name, value in env_items.items():
     command += ["--env", f"{name}={value}"]
-command += [SERVER_NAME, "--", str(Path(sys.executable).resolve()), "-m", "mcp-server"]
+command += [SERVER_NAME, "--", str(Path(sys.executable).resolve()), "-m", SERVER_NAME]
 
 subprocess.run(command, check=True)
 
@@ -79,7 +77,6 @@ codex_config.write_text(tomlkit.dumps(config), encoding="utf-8")
 print(f"Installed {SERVER_NAME} with {codex}")
 print(f"Python: {Path(sys.executable).resolve()}")
 print(f"Working directory: {repo}")
-print(f"PYTHONPATH: {repo}")
 print(f"Codex config: {codex_config}")
 print(f"Environment variables registered with codex --env: {len(env_items)}")
 print(f"tool_timeout_sec: {args.timeout}")

package/metaskills/academic-army-coding-plan/ENVOLVETASK.md

@@ -1 +1 @@
-Create a coding plan based on the paper blueprint in workspace/paper_blueprint.md and the experiment plan in workspace/experiment_plan.md.
+Create a coding plan based on the paper blueprint in workspace/plan/paper_blueprint.md and the experiment plan in workspace/plan/experiment_plan.md.

package/metaskills/academic-army-experiment-plan/ENVOLVETASK.md

@@ -1 +1 @@
-Create an experiment plan based on the paper blueprint in workspace/paper_blueprint.md.
+Create an experiment plan based on the paper blueprint in workspace/plan/paper_blueprint.md.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "academic-army",
-  "version": "0.4.0",
+  "version": "0.5.0",
   "description": "Agent workflows and skills for AcademicArmy.",
   "license": "MIT",
   "repository": {
@@ -23,10 +23,8 @@
     "skills",
     "metaskills",
     "runs",
-    "mcp-server/__main__.py",
-    "mcp-server/requirements.txt",
-    "mcp-server/deepresearch/*.py",
-    "mcp-server/writing_master/*.py",
+    "academic_army_mcp_tools/**/*.py",
+    "pyproject.toml",
     "install_mcp.py",
     "README.md",
     "README.zh-CN.md",
@@ -47,8 +45,8 @@
     "evolve-skill": "tsx src/cli.ts evolve-skill"
   },
   "dependencies": {
-    "coding-agent-forge": ">=1.3.5",
-    "developing-agent-forge": ">=2.8.3"
+    "coding-agent-forge": ">=1.5.0",
+    "developing-agent-forge": ">=2.9.0"
   },
   "devDependencies": {
     "@eslint/js": "^10.0.1",