mirror of
https://github.com/deepseek-ai/deepseek-harness
synced 2026-08-15 21:04:50 +00:00
66e56bd395
Codex review of PR #88 found three issues in the example-app extraction: A1 — examples/coding-agent/README.md's plugin table still listed the OLD direct-wired leaf entries (agent-loop, session-persistence, src/stdio-chat.ts — the whole src/ dir is gone). Rewrite it to the four real leaf entries the current cordis.yml loads (hmr, llm-deepseek, bash, stdio-agent), noting that tool-bash/persistence/agent/loop now live inside the agent-core + stdio-agent bundles. A2 — the three new app/spine packages (agent-core, stdio-agent, acp-agent) export NO `inject`, so a stray `export default apply` would let unwrapExports collapse the module and silently DROP name/Config WITHOUT crashing — the real-load-path smokes would stay green. agent-core is never Loader-unwrapped at all. Add an explicit export-shape guard per package: assert no `default` export and that the real Loader.unwrapExports leaves name/Config/apply intact. Verified each fails when `export default apply` is added. B — soften "structurally unreachable / cannot wire a stdout logger" overclaims in the acp-agent/agent-core READMEs and the implemented RFC: a leaf CAN still add a sibling logger entry; the accurate claim is the app omits one so the default leaf has nothing to get wrong. Keep the safety directive (never add a stdout logger to an ACP leaf).
40 lines
3.5 KiB
Markdown
40 lines
3.5 KiB
Markdown
# acp-agent example
|
|
|
|
The DeepSeek Harness coding agent exposed as an **Agent Client Protocol (ACP)** server over JSON-RPC stdio — drive it from Zed or any other ACP client.
|
|
|
|
```sh
|
|
pnpm run demo:acp # needs DEEPSEEK_API_KEY (repo-root .env or env)
|
|
```
|
|
|
|
This example is just a leaf `cordis.yml`: it loads the [`@deepseek-ai/dsh-acp-agent`](../../packages/ui/acp-agent) app (which bundles the [`@deepseek-ai/dsh-agent-core`](../../packages/core/agent-core) spine, JSONL session persistence, and the `@deepseek-ai/dsh-acp` bridge — with **no pre-created agents**, since ACP `session/new` creates them on demand) plus the two swappable backends (`llm-deepseek`, `bash-local`). The app package bakes in the no-stdout-logger cluster, so a leaf has no logger entry to get wrong by default — keeping stdout pure for JSON-RPC.
|
|
|
|
## stdout is the protocol
|
|
|
|
This example loads **no stdout logger** — `stdout` carries the JSON-RPC frames, and any other write corrupts them. `@deepseek-ai/dsh-acp-agent` includes no logger entry, so this leaf has none to get wrong by default; do not add one (use a stderr exporter if you need logs).
|
|
|
|
## Zed configuration
|
|
|
|
Add to your Zed `settings.json` under `agent_servers`:
|
|
|
|
```json
|
|
{
|
|
"agent_servers": {
|
|
"DeepSeek Harness": {
|
|
"command": "pnpm",
|
|
"args": ["--dir", "/path/to/deepseek-harness", "run", "demo:acp"],
|
|
"env": { "DEEPSEEK_API_KEY": "sk-…" }
|
|
}
|
|
}
|
|
}
|
|
```
|
|
|
|
The editor sets each session's `cwd` to the project it opens; the agent's bash tools run there (see the per-session `cwd` note in `packages/ui/acp`), so launch the server from the harness repo with `pnpm --dir …` and let ACP carry the workspace path per session.
|
|
|
|
## Snapshot tests (record-once / replay-deterministic)
|
|
|
|
This example is the home of the harness's **snapshot tests** — they boot this server as a real subprocess, drive it with a deterministic input script, and diff its normalized output against committed golden files. The model is made deterministic by `@deepseek-ai/dsh-llm-replay`, a function/namespace plugin that installs an `llm/stream` waterfall listener and short-circuits it, serving model streams reconstructed from a recorded **session JSONL** fixture (`<scenario>/session.jsonl`) — so replay needs no API key. The fixture IS the persisted session log: its `assistant/chunk` events carry every `StreamChunk`, so grouping them by `(turn, step)` reconstructs each `stream()` call (one model call per loop step). Recording is therefore "run the real agent once and harvest the `.jsonl`". The two failure modes not expressible as logged chunks — a pure throw before any chunk, and cancel/hang — use an optional `<scenario>/replay.override.json` sidecar (a `ReplayEntry[]` that replaces the derived script). A scenario that needs the agent to operate on existing files ships an optional `<scenario>/workspace/` directory — the harness copies its contents into the temp cwd before the run (see `workspace-edit`). See [docs/rfc/implemented/2026-06-19-acp-snapshot-tests.md](../../docs/rfc/implemented/2026-06-19-acp-snapshot-tests.md) for the full design.
|
|
|
|
## MVP limitations
|
|
|
|
The bridge supports N concurrent sessions per connection, each in its own workspace `cwd` (RFC 011). Remaining limits: prompts support ACP's baseline `text` and `resource_link` blocks only, `additionalDirectories` and `mcpServers` are rejected, and the tool-permission gate is deferred (`TODO(rfc010-permission-gate)` — tools run with the executor's full authority). See `packages/ui/acp/README.md` for the full contract.
|