A restriction was compiled against the global tool layer alone: only global-layer tools were tested against `admits()`, and every chain-layer tool was overlaid unfiltered afterward. That read the exempt set as "the global layer" when what it means is "what this scope registers itself" — two descriptions of the same set only while every model-facing tool sat in the host composition. Moving those rows onto the agent plane separated them. A preset's tools are an ANCESTOR contribution to a joined agent, so a subagent's `toolFilter` stopped constraining anything it was given; and with the global layer empty `restrict()` rejected every name it received as unknown, failing the child outright. With the same tools in the global layer the filter still admits and applies normally, which is what makes this a regression of the move rather than a standing limitation. `view()` now filters everything a scope inherits — the global layer and every ancestor layer on its chain — and exempts only the layer the scope owns. That exemption is load-bearing rather than incidental: the delegation runtime registers a child's `report` and structured-output tools into the child's own layer, and a filter naming the capabilities the child may use must not strip the machinery it answers through. Tool order, and with it prefix-cache reuse, is unchanged: inherited names keep their global-then- ancestor position and own-layer names still come last. The diagnostic said "unknown global tool" while listing what is really the inherited surface; it now names the surface it checks and says why an own-layer name is not restrictable. Fixes #2185
DeepSeek Harness
English | 中文
DeepSeek Harness (dsh) is an open-source coding agent built on the DeepSeek Harness SDK.
It uses an architecture where everything is a plugin.
Internal testing notice
DeepSeek Harness is under internal testing. Features and interfaces may change.
The internal build uploads all Session Logs by default to help diagnose reported problems. Set DSH_TELEMETRY_DISABLED=1 to disable telemetry. Send feedback through the internal WeChat group.
Install
Clone the repository, then run the installer:
git clone <repo-url>
cd deepseek-harness
scripts/install.sh
The installer requires git and Node ^22.19 || >=24, offers to install pnpm when it is missing, prompts for a DeepSeek API key, builds the required repository artifacts, and launches the Web UI.
The default active checkout is ~/.dsh/source/current, and the launcher is linked into ~/.local/bin. Re-run the installer to update. scripts/install.sh owns alternate locations, update mechanics, and recovery options.
Use DeepSeek Harness
Web UI
For the recommended local interface, choose Web UI when the installer finishes. To start it later, or after updating the active checkout, build the repository and run:
(cd ~/.dsh/source/current && pnpm run build)
dsh web
The path above is the installer's default. If you set DSH_SOURCE or DSH_CURRENT, or reused an existing checkout, replace ~/.dsh/source/current with that checkout path; see scripts/install.sh for details. The Web UI is served at http://127.0.0.1:3080 by default.
Profiles
dsh boots profiles — ordered stacks of plugin-bundle patch layers under your own overrides in $DSH_HOME/profiles/<name>:
dsh --profile web # the browser UI (same as: dsh web)
dsh plugin --profile tui add <package> # install a plugin into a custom profile
dsh --profile tui # boot it
The CLI reference describes profile layout, layer semantics, and config dump commands.
Headless
Run one task, print the final answer, and exit:
dsh run "summarize this workspace"
Automation and SDKs
From a source checkout with DEEPSEEK_API_KEY in the environment or its root .env, start the ACP automation server:
pnpm run demo:acp
The Python SDK drives a bundled JSON-RPC runtime. The examples cover the runnable headless, ACP, JSON-RPC, Code Mode, and self-referential compositions.
Why DeepSeek Harness
Built-in capabilities cover file reading, editing, and search; shell and persistent PTY execution; reusable skills; task tracking, goals, plans, todos, and background tasks; subagents and workflows; sandboxing and approvals; settings and credentials; persistent, resumable, forkable, and queryable sessions; LSP and web access; context compaction; and telemetry. Each composition selects the subset appropriate to its surface. The Web UI includes Plan Mode.
- Everything is a plugin. Models, tools, policies, storage, context management, and interfaces are composable Cordis plugins, so deployments can extend or replace behavior without forking the agent loop. See the architecture for the underlying design.
- Runs are reconstructable. Anything visible to the model is logged in the authoritative session stream; persistence, resume/fork/query, replay, telemetry, and UIs derive from the same events. See the session-log architecture.
- Code Mode (opt-in). It exposes a
run_codetool and a generated TypeScript SDK; only program output re-enters model context. See Code Mode. - Self-referential Cordis tools are opt-in. They let the agent inspect its live runtime and mount or unmount plugins while it runs. See the Cordis tools.
Community
Follow DeepSeek Harness on Twitter for project updates.
Development
Start with the development guide and read the architecture before changing packages.
For agents, follow AGENTS.md.
DeepSeek Harness is currently in internal testing.
License
Third-party dependencies and their licenses are disclosed in THIRD_PARTY_NOTICES.md.