Coder/deepseek-harness
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: retain JSON-RPC subagent locality per run

Browse Source
This commit is contained in:
Tianyi Cui
2026-07-14 08:11:13 +08:00
parent ff4636e9d9
commit cb80277e60
4 changed files with 98 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
packages/ui/jsonrpc/README.md
View File

@@ -4,7 +4,7 @@ The **SDK server plugin** (`jsonrpc`): mounting it serves a stdio JSON-RPC serve
## Wiring
`inject: ['agents']` — the server creates one agent per SDK `sessionId` (get-or-create on `session/prompt`). A local subagent's shared agent/session id supplies `subagent.finished.childSessionId` directly; the server caches runtime-local identity plus optional parent lineage because the child may be disposed before `subagent/end`, and the provider contract does not require lineage. Runs from remote providers are not reported because they create no local agent. The LLM seam is read opportunistically via `ctx.get('llm')` (not injected): when `initialize.model` has no registered adapter, the plugin mounts `dsh-llm-deepseek` for it (credentials from `$DEEPSEEK_API_KEY` / `$DEEPSEEK_BASE_URL`) — a config-registered adapter for the model wins. Everything else — persistence, the tool stacks, the adapter set — comes from the surrounding `cordis.yml`.
`inject: ['agents']` — the server creates one agent per SDK `sessionId` (get-or-create on `session/prompt`). A local subagent's shared agent/session id supplies `subagent.finished.childSessionId` directly; the server caches runtime-local identity plus optional parent lineage for the agent lifetime and snapshots it per run, because a continuation may reuse one child and the child may be disposed before a later `subagent/end`. Runs from remote providers are not reported because they create no local agent. The LLM seam is read opportunistically via `ctx.get('llm')` (not injected): when `initialize.model` has no registered adapter, the plugin mounts `dsh-llm-deepseek` for it (credentials from `$DEEPSEEK_API_KEY` / `$DEEPSEEK_BASE_URL`) — a config-registered adapter for the model wins. Everything else — persistence, the tool stacks, the adapter set — comes from the surrounding `cordis.yml`.
## Config