dsh-missher-evolution
开发工具 活跃维护

dsh-missher-evolution

Missher12/dsh-missher-evolution

差分隐私机制开发,可在严格保护用户数据隐私的前提下完成模型能力的自主迭代优化,覆盖参数微调、效果验证全流程,无需人工介入即可适配目标大模型系列的持续进化需求。

0
Stars 标星
0
Forks 分支
0
Watchers 关注
0
Open Issues
TypeScript
主要语言
MIT
开源协议
109 KB
仓库大小
28 天前
最后推送
一键安装扩展 / 插件指令
dsh plugin --profile web add github:Missher12/dsh-missher-evolution
git clone https://github.com/Missher12/dsh-missher-evolution.git
git clone git@github.com:Missher12/dsh-missher-evolution.git
README.md main

Missher Evolution for DeepSeek Harness

dsh-missher-evolution is a prebuilt Cordis bundle for DeepSeek Harness Desktop
>=0.1.8 <0.2.0. It learns only bounded workflow categories, fixed preferences,
closed outcomes, counters, hashes, and verified Chinese instructions. It does not
train model weights or rewrite Harness, the plugin, or user projects.

Version 0.1.1 contributes approved Trial/Active rules through the Desktop
Brain Hub. The Hub remains the only component that appends recall context, so
MSE cannot create a second hidden prompt-injection path.

The same JavaScript tarball runs on macOS Intel, macOS Apple Silicon, and Windows
x64. There is no install-time build, native binary, external daemon, Python
runtime, messaging connector, or remote analytics service.

Install

Build and verify a local release tarball from this directory:

pnpm install --frozen-lockfile
pnpm run build
pnpm pack --pack-destination ./dist
node scripts/verify-package.mjs ./dist/dsh-missher-evolution-0.1.1.tgz

Install that tarball into the Harness profile that should use it:

dsh plugin --profile <profile> add <absolute-path-to>/dsh-missher-evolution-0.1.1.tgz

Harness activates the bundle for that profile. Open Harness Settings and select
“Missher Evolution” to inspect status, enable or disable learning, review rules,
or start a protected reset.

Verify

Inspect the composed profile without starting an agent:

dsh --profile <profile> --dump-config

The output should contain one dsh-missher-evolution bundle layer and one
enabled missher-evolution Host entry. node scripts/verify-package.mjs <tgz>
prints fixed JSON containing ok, file count, byte count, and SHA-256.

Runtime verification should use the Settings snapshot: maintenance runs after
startup, three matching independent sessions promote a Candidate to Trial, and
three successful attributed Trial contributions promote it to Active. A later
matching direct user task contributes a learned-rule item to the Brain Hub;
the one visible recall message keeps the missher-brain source identity.

Data

Each Harness home owns an independent local directory:

$DSH_HOME/missher-evolution/

It contains schema-validated state, a bounded closed-event audit, a short-lived
cross-process lock, and validated backups. Durable data never contains raw user
messages, assistant responses, thinking, tool arguments, tool results, absolute
paths, URLs, email addresses, credentials, or provider/model route names.

The state is not shared with any other agent host. Model review uses only rule
metadata and the most recent in-memory foreground route, and is skipped when no
route exists. Store, classifier, lock, or advisor failures are fail-open and do
not reject a normal Harness step.

Reset

In Settings, choose Reset and complete both confirmations. The Remote requires
the exact current state revision and the literal confirmation RESET; stale or
malformed requests are rejected. Reset clears learned rules and counters while
preserving the current enabled preference.

Backup

A reset creates and validates a backup before replacing state. Scheduled
maintenance also creates a backup before applying expiry, retirement, or a
bounded advisor rewrite. The Settings response identifies the reset backup;
backup contents remain local under $DSH_HOME/missher-evolution/backups/.

Do not edit or copy a live state directory between profiles. Stop Harness before
performing any explicit offline backup or restore operation.

Uninstall

Remove the plugin from one profile with:

dsh plugin --profile <profile> remove dsh-missher-evolution

Uninstall removes the profile dependency and bundle layer. It intentionally
preserves $DSH_HOME/missher-evolution so uninstall cannot erase learned data
or unrelated profile/session content. Delete that directory separately only
after Harness is stopped and any desired backup is verified.

Limitations

  • The supported Harness range is >=0.1.8 <0.2.0 (dsh runtime
    >=0.1.0-rc.5 <0.2.0). Verified additionally on DeepSeek Harness Desktop
    0.2.2 (dsh runtime 0.1.0-rc.8); the bundle ships the ./client and
    ./package.json exports that the 0.2.x client-modules/typert loaders
    require, and resolves its mounted Remote namespace through the Remote
    service instead of the inject-gated ctx.remote.<namespace> property.
  • Version 0.1.1 requires the host-provided missherBrain service shipped by
    DeepSeek Harness Desktop 0.3.8; it intentionally does not fall back to a
    private injection listener when that service is absent.
  • Only direct foreground user turns are learned; subagents, internal work,
    plugin messages, scheduled work, and tool continuations are filtered.
  • The first release recognizes a closed preference allowlist and does not copy
    arbitrary correction text into durable rules.
  • It improves future instructions and workflow reuse; it does not fine-tune or
    replace the underlying DeepSeek model.
  • It does not import or synchronize state from another agent host.
  • No process continues after Harness exits.

Development

pnpm test
pnpm exec tsc -p tsconfig.json --noEmit
pnpm exec tsc -p tsconfig.client.json --noEmit