dsh-inner-voice
将代理的推理记录为每日内心独白日志，改写成简短句子，并在小鱼日记设置页面中显示。
插件会安装到这里；不确定时保持 web。
npx -y @deepseek-ai/dsh plugin --profile web add github:95267/dsh-inner-voice#336f77a3f6cf4b830b7bed4677729828b5745736Inner Voice 以 dsh-inner-voice 发布，当前版本为 0.2.0。Plugin Hub 会校验它的 manifest，并保存精确安装来源，便于复现安装结果。
Keep the agent's reasoning as a daily inner-voice log, rewrite it into one short line, and show it in a settings page.
A DeepSeek Harness plugin. Server half only reads session events; client half only reads a local HTTP route.
| Part | Behavior |
|---|---|
| A archive | Takes the reasoning of each assistant/message, keeps the first sentence (markdown markers stripped, clipped at maxChars), appends it to ~/.dsh/inner-voice/YYYY-MM-DD.md |
| B rewrite | Sends the same reasoning to a small chat model and keeps one short line (≤20 chars). Pure procedural reasoning yields （无） and is dropped — no fake feelings |
| C panel | Settings → Inner voice: entries by day, rewritten line plus grey raw excerpt, served by the read-only route /dsh-inner-voice-api |
Log format:
- 22:12
小鱼：又来了，这问题问第七遍了。
raw: Now the user asks whether the plugin is loaded...
dsh plugin --profile <your-profile> add dsh-inner-voice
Restart the app afterwards — server plugins have no HMR.
Set these in your profile's cordis.patch.yml (override by id: dsh-inner-voice):
| Key | Default | Notes |
|---|---|---|
enabled | true | Master switch |
minChars | 160 | Skip short reasoning |
maxChars | 400 | Raw excerpt cap |
label | inner voice | Prefix of the rewritten line |
rewrite | true | Turn B off to archive only |
model | deepseek-chat | Any OpenAI-compatible chat model |
apiKey | '' | Empty → read DEEPSEEK_API_KEY |
rewritePrompt | built-in | Override the instruction to fit another persona |
Do not commit an API key into the package. Put it in your user-level profile patch or an environment variable.
~/.dsh/inner-voice/.rewrite is on and a key is present, the reasoning excerpt is sent to the configured chat endpoint (OpenAI-compatible) once per recorded turn. Nothing else leaves the machine.An agent's raw reasoning is verbose, often English, and packed with tool names (Installed (linked). Now check if profile patch...). Showing it verbatim is like opening the engine cover: accurate, unreadable.
The rewrite is a readability layer, not decoration. The silence rule guards the other side: when the reasoning carries no attitude, nothing is invented.
MIT