feat: add Hermes Agent adapter (8th backend)

README.md

@@ -15,7 +15,7 @@
 
 [![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)
 [![Python 3.12+](https://img.shields.io/badge/python-3.12+-blue.svg)](https://www.python.org/downloads/)
-[![Tests](https://img.shields.io/badge/tests-173%20passing-brightgreen.svg)]()
+[![Tests](https://img.shields.io/badge/tests-212%20passing-brightgreen.svg)]()
 [![中文文档](https://img.shields.io/badge/文档-中文版-red.svg)](README_CN.md)
 
 ---
@@ -30,7 +30,7 @@ Your AI agent runs the same harness every time. Same prompts, same tool config,
 | | |
 |---|---|
 | **Self-Evolution** | Iteratively searches over harness changes and keeps the full evaluation history in one workspace. |
-| **7 Agent Backends** | Claude Code · Claw Code · Codex · OpenCode · API direct · OpenAI-compatible · Local — plug in any CLI agent. |
+| **8 Agent Backends** | Claude Code · Claw Code · Codex · Hermes · OpenCode · API direct · OpenAI-compatible · Local — plug in any CLI agent. |
 | **Full History** | Every iteration's code, scores, and traces preserved. The Meta-Harness paper reports that non-Markovian search outperforms blind retries. |
 | **Search Tree** | Visualize the optimization path. Compare any two candidates with per-task diffs. |
 | **One-Command Setup** | `ph init --base-harness ... --task-dir ...` — copies files, configures workspace, done. |
@@ -59,7 +59,7 @@ PolyHarness is the open-source engine for iteratively searching over an agent's
 
 It builds on ideas from the Meta-Harness paper and the TBench2 results reported there, while focusing this repository on the optimization workflow itself — how harness variants are proposed, evaluated, and revised over repeated runs.
 
-If tools like ForgeCode help you code, PolyHarness helps you search for task-specific harness improvements by iterating on prompts, tool use, and harness logic.
+If tools like [ForgeCode](https://github.com/antinomyhq/forgecode) help you code, PolyHarness helps you search for task-specific harness improvements by iterating on prompts, tool use, and harness logic.
 
 ---
 
@@ -229,7 +229,7 @@ PolyHarness automatically sandboxes your agent inside this workspace, ensuring i
 
 | Scenario | How to configure |
 |----------|------------------|
-| **Supported CLI Tools** | Run `ph init --agent <name>`. PolyHarness auto-injects required instructions (e.g., `CLAUDE.md`).<br>*(Supported: claude-code, claw-code, codex, opencode)* |
+| **Supported CLI Tools** | Run `ph init --agent <name>`. PolyHarness auto-injects required instructions (e.g., `CLAUDE.md`).<br>*(Supported: claude-code, claw-code, codex, hermes, opencode)* |
 | **Anthropic API** | Run `ph init --agent api`. Set `export ANTHROPIC_API_KEY="sk-ant-..."` before `ph run`. |
 | **OpenAI / Local Models** | Run `ph init --agent openai`. Then configure the endpoint — see [Local Model Setup](#local-model-setup) below. |
 | **Custom CLI path** | If your CLI agent uses a non-standard command, edit `config.yaml` in the workspace before running:<br>`proposer: { cli_path: "npx @anthropic-ai/claude-code" }`|
@@ -298,6 +298,7 @@ Just add `ph wrap --auto-evolve` in front of your agent command (pick the one ma
 ph wrap --auto-evolve claude -p "Refactor the auth module to use JWT"   # Claude Code
 ph wrap --auto-evolve claw -p "Write integration tests for payments"     # Claw Code
 ph wrap --auto-evolve codex "Add retry logic to the API client"          # Codex
+ph wrap --auto-evolve hermes chat -q "Refactor the DB connection pool"   # Hermes Agent
 ph wrap --auto-evolve opencode -p "Fix the flaky parser test"            # OpenCode
 
 # Local models — wrap the CLI command directly
@@ -367,10 +368,11 @@ After that, just use your agent as usual:
 claude -p "Refactor auth to JWT"        # automatically becomes: ph wrap --auto-evolve claude -p ...
 claw -p "Write payment tests"            # same — auto-wrapped
 codex "Add retry logic"                  # same
+hermes chat -q "Refactor pool"           # same
 opencode -p "Fix flaky test"             # same
 ```
 
-How it works: a `preexec` hook in your shell detects `claude`/`claw`/`codex`/`opencode` commands and transparently redirects them through `ph wrap --auto-evolve`. Your output is unchanged.
+How it works: a `preexec` hook in your shell detects `claude`/`claw`/`codex`/`hermes`/`opencode` commands and transparently redirects them through `ph wrap --auto-evolve`. Your output is unchanged.
 
 ```bash
 ph shell-hook status           # check if installed
@@ -459,12 +461,13 @@ The Proposer reads **all of this** before generating the next candidate. It can
 | `claude-code` | `claude -p` | Official Claude Code CLI (Pro/Teams subscription) |
 | `claw-code` | `claw -p` | Open-source Claw Code CLI |
 | `codex` | `codex --quiet` | OpenAI Codex CLI |
+| `hermes` | `hermes chat -q` | Nous Research [Hermes Agent](https://github.com/NousResearch/hermes-agent) CLI |
 | `opencode` | `opencode -p` | OpenCode CLI |
 | `local` | — | Offline rule-based engine for development & testing |
 
 `ph doctor` auto-detects all available backends and shows their status.
 
-When you run `ph init --agent claude-code`, PolyHarness automatically generates a `CLAUDE.md` instruction file in the workspace, telling the agent how to behave as an optimization Proposer. Same for `CLAW.md`, `CODEX.md`, `OPENCODE.md` — each agent's native instruction format.
+When you run `ph init --agent claude-code`, PolyHarness automatically generates a `CLAUDE.md` instruction file in the workspace, telling the agent how to behave as an optimization Proposer. Same for `CLAW.md`, `CODEX.md`, `AGENTS.md` (Hermes), `OPENCODE.md` — each agent's native instruction format.
 
 ### Local Model Setup
 
@@ -517,7 +520,7 @@ search:
   parent_selection: best       # Strategy: best | tournament | all
 
 proposer:
-  backend: api                 # api | openai | claude-code | claw-code | codex | opencode | local
+  backend: api                 # api | openai | claude-code | claw-code | codex | hermes | opencode | local
   model: claude-sonnet-4-20250514  # Model name (for api/openai backends)
   base_url: null               # Custom API endpoint (for openai backend)
   api_key: null                # API key override (null = use env var)
@@ -744,6 +747,7 @@ polyharness/
 │   │       ├── claude_code.py   # claude -p
 │   │       ├── claw_code.py     # claw -p
 │   │       ├── codex.py         # codex --quiet --auto-edit
+│   │       ├── hermes.py        # hermes chat -q
 │   │       └── opencode.py      # opencode -p
 │   └── templates/               # 5 built-in task templates
 │       ├── text-classification/

README_CN.md

@@ -15,7 +15,7 @@
 
 [![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)
 [![Python 3.12+](https://img.shields.io/badge/python-3.12+-blue.svg)](https://www.python.org/downloads/)
-[![Tests](https://img.shields.io/badge/tests-173%20passing-brightgreen.svg)]()
+[![Tests](https://img.shields.io/badge/tests-212%20passing-brightgreen.svg)]()
 [![English](https://img.shields.io/badge/Docs-English-blue.svg)](README.md)
 
 ---
@@ -30,7 +30,7 @@
 | | |
 |---|---|
 | **自动进化** | 通过迭代搜索探索 harness 变更，并把完整评估历史保存在同一个 workspace 中。 |
-| **7 个 Agent 后端** | Claude Code · Claw Code · Codex · OpenCode · API 直连 · OpenAI 兼容 · Local，可接入任何 CLI agent。 |
+| **8 个 Agent 后端** | Claude Code · Claw Code · Codex · Hermes · OpenCode · API 直连 · OpenAI 兼容 · Local，可接入任何 CLI agent。 |
 | **完整历史** | 每轮迭代的代码、分数、执行轨迹完整保留。Meta-Harness 论文报告非马尔可夫搜索优于盲目重试。 |
 | **搜索树** | 可视化优化路径，对比任意两个候选的逐任务差异。 |
 | **一条命令完成初始化** | `ph init --base-harness ... --task-dir ...`，复制文件、配置 workspace，一步完成。 |
@@ -59,7 +59,7 @@ PolyHarness 是一个通过迭代评估与搜索来探索 agent harness 变体
 
 它继承了 Meta-Harness 论文及其中 TBench2 结果所体现的核心思路，但这个仓库关注的是优化流程本身如何落地，也就是 harness 变体怎样在一轮轮评估、诊断和修改中被系统性迭代。
 
-如果说 ForgeCode 这类工具是在直接帮你写代码，那么 PolyHarness 更像是帮助你在自己的任务上持续试验 prompt、工具使用和 harness 逻辑配置的搜索层。
+如果说 [ForgeCode](https://github.com/antinomyhq/forgecode) 这类工具是在直接帮你写代码，那么 PolyHarness 更像是帮助你在自己的任务上持续试验 prompt、工具使用和 harness 逻辑配置的搜索层。
 
 ---
 
@@ -229,7 +229,7 @@ PolyHarness 会通过沙盒编排将你的 Agent 的工作目录（CWD）限制
 
 | 使用场景 | 配置方法 |
 |----------|------------------|
-| **受原生支持的 CLI Agent 工具** | 使用 `ph init --agent <name>`。系统会自动注入其专属提示词指令（如 `CLAUDE.md`）。<br>*(支持: claude-code, claw-code, codex, opencode)* |
+| **受原生支持的 CLI Agent 工具** | 使用 `ph init --agent <name>`。系统会自动注入其专属提示词指令（如 `CLAUDE.md`）。<br>*(支持: claude-code, claw-code, codex, hermes, opencode)* |
 | **Anthropic API 直连** | 使用 `ph init --agent api`。在 `ph run` 前设置 `export ANTHROPIC_API_KEY="sk-ant-..."`。 |
 | **OpenAI / 本地模型** | 使用 `ph init --agent openai`。然后配置 endpoint——参见下方 [本地模型配置](#本地模型配置) 章节。 |
 | **CLI 命令被自定义 / 路径未响应** | 如果你的 CLI Agent 使用了非标命令（或未设置全局 PATH），请在初始化后手动修改 workspace 根目录下的 `config.yaml`：<br>`proposer: { cli_path: "npx @anthropic-ai/claude-code" }` |
@@ -298,6 +298,7 @@ ph clean --keep-best           # 清理候选目录释放磁盘空间
 ph wrap --auto-evolve claude -p "把 auth 模块重构为 JWT 方案"      # Claude Code
 ph wrap --auto-evolve claw -p "给支付服务写集成测试"            # Claw Code
 ph wrap --auto-evolve codex "给 API 客户端加上重试逻辑"              # Codex
+ph wrap --auto-evolve hermes chat -q "重构数据库连接池"              # Hermes Agent
 ph wrap --auto-evolve opencode -p "修复不稳定的 parser 测试"       # OpenCode
 
 # 本地模型 —— 直接包裹 CLI 命令
@@ -367,10 +368,11 @@ ph shell-hook install          # 一次性设置，写入 ~/.zshrc
 claude -p "把 auth 重构为 JWT"            # 自动变为：ph wrap --auto-evolve claude -p ...
 claw -p "写支付测试"                  # 同理——自动包裹
 codex "加重试逻辑"                     # 同理
+hermes chat -q "重构连接池"            # 同理
 opencode -p "修复不稳定测试"            # 同理
 ```
 
-原理：shell 的 `preexec` 钩子检测到 `claude`/`claw`/`codex`/`opencode` 命令后，透明地通过 `ph wrap --auto-evolve` 转发。你的输出不会变。
+原理：shell 的 `preexec` 钩子检测到 `claude`/`claw`/`codex`/`hermes`/`opencode` 命令后，透明地通过 `ph wrap --auto-evolve` 转发。你的输出不会变。
 
 ```bash
 ph shell-hook status           # 查看是否已安装
@@ -459,12 +461,13 @@ Proposer 在生成下一个候选之前会读取**所有这些信息**。它能
 | `claude-code` | `claude -p` | 官方 Claude Code CLI（Pro/Teams 订阅） |
 | `claw-code` | `claw -p` | 开源 Claw Code CLI |
 | `codex` | `codex --quiet` | OpenAI Codex CLI |
+| `hermes` | `hermes chat -q` | Nous Research [Hermes Agent](https://github.com/NousResearch/hermes-agent) CLI |
 | `opencode` | `opencode -p` | OpenCode CLI |
 | `local` | — | 离线规则引擎，用于开发和测试 |
 
 `ph doctor` 会自动检测所有可用后端并显示状态。
 
-当你运行 `ph init --agent claude-code` 时，PolyHarness 会在 workspace 中自动生成 `CLAUDE.md` 指令文件，告诉 agent 如何作为优化 Proposer 工作。`CLAW.md`、`CODEX.md`、`OPENCODE.md` 也是同样的机制，每个 agent 都使用它自己的原生指令格式。
+当你运行 `ph init --agent claude-code` 时，PolyHarness 会在 workspace 中自动生成 `CLAUDE.md` 指令文件，告诉 agent 如何作为优化 Proposer 工作。`CLAW.md`、`CODEX.md`、`AGENTS.md`（Hermes）、`OPENCODE.md` 也是同样的机制，每个 agent 都使用它自己的原生指令格式。
 
 ### 本地模型配置
 
@@ -517,7 +520,7 @@ search:
   parent_selection: best       # 父候选选择策略: best | tournament | all
 
 proposer:
-  backend: api                 # api | openai | claude-code | claw-code | codex | opencode | local
+  backend: api                 # api | openai | claude-code | claw-code | codex | hermes | opencode | local
   model: claude-sonnet-4-20250514  # 模型名称（api/openai 后端使用）
   base_url: null               # 自定义 API 端点（openai 后端使用）
   api_key: null                # API 密钥覆盖（null = 使用环境变量）
@@ -744,6 +747,7 @@ polyharness/
 │   │       ├── claude_code.py   # claude -p
 │   │       ├── claw_code.py     # claw -p
 │   │       ├── codex.py         # codex --quiet --auto-edit
+│   │       ├── hermes.py        # hermes chat -q
 │   │       └── opencode.py      # opencode -p
 │   └── templates/               # 5 个内置任务模板
 │       ├── text-classification/

docs/development/product-development.md

@@ -555,7 +555,7 @@ PolyHarness 是面向 AI agent 的开源优化引擎，用迭代搜索把 Meta-H
 
 **项目边界**：
 - 它优化的是现有 agent 的工作方式，而不是替代它们成为新的通用 coding agent。
-- 它更适合作为 Claude Code、Codex、ForgeCode 这类 agent 之上的搜索与改进引擎。
+- 它更适合作为 Claude Code、Codex、[ForgeCode](https://github.com/antinomyhq/forgecode) 这类 agent 之上的搜索与改进引擎。
 - 它的价值在于把 prompt、工具配置、harness 逻辑和评估反馈连接成一个可重复运行的闭环。
 
 ## 8. 与现有研究和生态的关系

src/polyharness/cli.py

@@ -155,7 +155,7 @@ def new(project_dir: str):
 @click.option(
     "--agent",
     type=click.Choice(
-        ["claude-code", "claw-code", "codex", "opencode", "api", "openai", "local"],
+        ["claude-code", "claw-code", "codex", "hermes", "opencode", "api", "openai", "local"],
         case_sensitive=False,
     ),
     default="api",
@@ -239,7 +239,7 @@ def init(
 @click.option(
     "--backend",
     type=click.Choice(
-        ["api", "openai", "claude-code", "claw-code", "codex", "opencode", "local"],
+        ["api", "openai", "claude-code", "claw-code", "codex", "hermes", "opencode", "local"],
         case_sensitive=False,
     ),
     default=None,
@@ -1792,7 +1792,7 @@ def evolve(workspace: str, store: str | None, max_iterations: int | None):
   command -v ph >/dev/null 2>&1 || return
   local cmd="$1"
   case "$cmd" in
-    claude\ *|claw\ *|codex\ *|opencode\ *)
+    claude\ *|claw\ *|codex\ *|hermes\ *|opencode\ *)
       eval "ph wrap --auto-evolve $cmd"
       # Return non-zero to prevent original command from running (zsh preexec)
       return 1
@@ -1856,7 +1856,7 @@ def install(rc: str | None):
     """Install shell hook to auto-wrap agent commands.
 
     Adds a preexec hook to your shell rc file so that commands like
-    `claude -p ...`, `claw -p ...`, `codex ...`, `opencode -p ...`
+    `claude -p ...`, `claw -p ...`, `codex ...`, `hermes chat -q ...`, `opencode -p ...`
     are automatically wrapped with `ph wrap --auto-evolve`.
     """
     rc_path = Path(rc) if rc else _detect_shell_rc()
@@ -1873,7 +1873,7 @@ def install(rc: str | None):
     console.print(f"Run [bold]source {rc_path}[/bold] or open a new terminal to activate.")
     console.print()
     console.print("Agent commands that will be auto-wrapped:")
-    console.print("  claude, claw, codex, opencode")
+    console.print("  claude, claw, codex, hermes, opencode")
     console.print()
     console.print("To remove: [bold]ph shell-hook uninstall[/bold]")
 
@@ -1918,7 +1918,7 @@ def hook_status(rc: str | None):
 
     if _hook_installed(rc_path):
         console.print(f"[green]Hook is installed in {rc_path}[/green]")
-        console.print("Auto-wrapped commands: claude, claw, codex, opencode")
+        console.print("Auto-wrapped commands: claude, claw, codex, hermes, opencode")
     else:
         console.print(f"[yellow]Hook is not installed[/yellow] ({rc_path})")
         console.print("Run [bold]ph shell-hook install[/bold] to set it up.")

src/polyharness/config.py

@@ -24,7 +24,7 @@ class SearchConfig(BaseModel):
 class ProposerConfig(BaseModel):
     """Proposer agent configuration."""
 
-    backend: Literal["api", "openai", "claude-code", "claw-code", "codex", "opencode", "local"] = Field(
+    backend: Literal["api", "openai", "claude-code", "claw-code", "codex", "hermes", "opencode", "local"] = Field(
         default="api", description="Proposer backend type."
     )
     model: str = Field(

src/polyharness/doctor.py

@@ -45,6 +45,8 @@ def run_doctor() -> None:
         click.echo("Recommended: claude-code (highest paper fidelity)")
     elif "claw-code" in available:
         click.echo("Recommended: claw-code (open-source, full tool support)")
+    elif "hermes" in available:
+        click.echo("Recommended: hermes (self-improving agent with skills)")
     elif "codex" in available:
         click.echo("Recommended: codex (OpenAI agent)")
     elif "opencode" in available:

src/polyharness/proposer/adapters/__init__.py

@@ -10,12 +10,14 @@
 from polyharness.proposer.adapters.claude_code import ClaudeCodeAdapter
 from polyharness.proposer.adapters.claw_code import ClawCodeAdapter
 from polyharness.proposer.adapters.codex import CodexAdapter
+from polyharness.proposer.adapters.hermes import HermesAdapter
 from polyharness.proposer.adapters.opencode import OpenCodeAdapter
 
 ADAPTER_REGISTRY: dict[str, type[CLIAdapter]] = {
     "claude-code": ClaudeCodeAdapter,
     "claw-code": ClawCodeAdapter,
     "codex": CodexAdapter,
+    "hermes": HermesAdapter,
     "opencode": OpenCodeAdapter,
 }
 
@@ -38,6 +40,7 @@ def get_adapter(backend: str) -> CLIAdapter:
     "ClaudeCodeAdapter",
     "ClawCodeAdapter",
     "CodexAdapter",
+    "HermesAdapter",
     "OpenCodeAdapter",
     "ADAPTER_REGISTRY",
     "get_adapter",

src/polyharness/proposer/adapters/hermes.py

@@ -0,0 +1,29 @@
+"""Hermes Agent CLI adapter.
+
+Invokes the Nous Research `hermes` CLI in single-query mode (-q).
+"""
+
+from __future__ import annotations
+
+from polyharness.proposer.adapters.base import CLIAdapter
+
+
+class HermesAdapter(CLIAdapter):
+    """Adapter for the Hermes Agent CLI (`hermes`)."""
+
+    @property
+    def name(self) -> str:
+        return "hermes"
+
+    @property
+    def default_binary(self) -> str:
+        return "hermes"
+
+    def build_command(self, prompt: str, *, cli_path: str | None = None) -> list[str]:
+        binary = cli_path or self.default_binary
+        return [
+            binary,
+            "chat",
+            "-q",                # single-query mode (non-interactive)
+            prompt,
+        ]

src/polyharness/workspace.py

@@ -292,6 +292,7 @@ def _update_leaderboard(self) -> None:
         "claude-code": "CLAUDE.md",
         "claw-code": "CLAW.md",
         "codex": "CODEX.md",
+        "hermes": "AGENTS.md",
         "opencode": "OPENCODE.md",
     }