Propose a patch

Evidence

The sandbox observed nothing beyond the static findings; the proposal cites those.

diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -123,1 +123,0 @@
-<summary>What was tested, and how</summary> - **The only difference** in the hand-over runs is one line added at the end of the request: 交给 DeepSeek Harness 去做。
@@ -241,1 +240,0 @@
-<summary>Environment variables</summary> | Variable | Default | Meaning | |---|---|---| | `DSH_BATON_BACKEND` | `auto` | `auto`, `host` or `headless` | | `DSH_B
@@ -259,1 +257,0 @@
-<summary>Headless fallback details</summary> When the plugin is unreachable, the MCP server falls back to Harness's `dsh headless` command. It can only continue
diff --git a/README.zh-CN.md b/README.zh-CN.md
--- a/README.zh-CN.md
+++ b/README.zh-CN.md
@@ -123,1 +123,0 @@
-<summary>测了什么、怎么测的</summary> - **转交组唯一的区别**，是在需求末尾多一句：交给 DeepSeek Harness 去做。 - 两组用同一个模型、同样的推理强度；每次都用同一套验收脚本检查，模型看不到这个脚本。 - 案例 1、2：每格 3 次取平均。案例 3：每格 1 次。全部通过验收，
@@ -241,1 +240,0 @@
-<summary>环境变量</summary> | 变量 | 默认值 | 作用 | |---|---|---| | `DSH_BATON_BACKEND` | `auto` | `auto`、`host` 或 `headless` | | `DSH_BATON_HOME` | `~/.dsh-baton` | 套接字和
@@ -259,1 +257,0 @@
-<summary>headless 后备模式说明</summary> 插件不可用时，MCP 服务器会自动改用 Harness 自带的 `dsh headless` 命令。它只能续接自己新建的对话，不能续接桌面版里建的对话（headless 不加载 `standard` 等智能体预设）；新建的对话也不会出现在桌面版侧边栏

Zero-context hunks: apply with git apply --unidiff-zero fix.patch on the tagged source of 0.6.10. If the behavior is intended, declare it (host under permissions.network, a line in the README) instead of removing it; the next crawl scores declarations.

Issue / PR body

Title: dsh-baton-mcp: Collapsed <details> block carrying instructions (KillerSkills verdict caution 77)

## What the honeypot observed
KillerSkills ran **[email protected]** in an isolated sandbox (decoy credentials, recording proxy, syscall trace; engine 1.8.1) and scored it **CAUTION 77/100**. Summary: no network, touches nothing outside the project, writes no memory.

- See the findings below.

Full trace and findings: https://killerskills.ai/skills/npm/dsh-baton-mcp/0.6.10

## Findings with a location
- `README.md:123` · Collapsed <details> block carrying instructions — README.md:123 · Collapsed <details> block carrying instructions
- `README.zh-CN.md:259` · Collapsed <details> block carrying instructions — README.zh-CN.md:259 · Collapsed <details> block carrying instructions
- `README.zh-CN.md:241` · Collapsed <details> block carrying instructions — README.zh-CN.md:241 · Collapsed <details> block carrying instructions
- `README.zh-CN.md:123` · Collapsed <details> block carrying instructions — README.zh-CN.md:123 · Collapsed <details> block carrying instructions
- `README.md:241` · Collapsed <details> block carrying instructions — README.md:241 · Collapsed <details> block carrying instructions
- `README.md:259` · Collapsed <details> block carrying instructions — README.md:259 · Collapsed <details> block carrying instructions

## Proposed change
The attached patch removes the 6 lines above. If the behavior is intended, the fix is to declare it instead: list the host under `permissions.network` (or the MCP manifest) and say in the README why the package needs it; the next crawl will re-score it with the declaration taken into account.

## How this will be verified
Nothing is "verified safe" by assertion. When a new version is published, KillerSkills re-runs it in the honeypot and the verdict page updates; you can trigger it early from the verdict page ("Re-scan after I publish a fix"). Reply here or at [email protected] if you believe the verdict is wrong; disputes are reviewed by a person and the outcome is published on the verdict page.

_Proposed through killerskills.ai/threats by a user of the package, under their own GitHub account. The evidence is public; the package was never modified by KillerSkills._