mirror of
https://github.com/deepseek-ai/deepseek-harness
synced 2026-08-15 21:04:50 +00:00
Machine-produced by `pnpm run rescope-vendor --apply` plus the regeneration it prints: `pnpm install` for the lockfile, `pnpm run gen-third-party-notices`, `verify-translation-pairing --write` for the touched bilingual pairs, `gen-doc-graphs`, and one typert snapshot whose ids embed character offsets. `pnpm run rescope-vendor --check` verifies the result. Renames nine vendored packages (cordis, cosmokit, schemastery and the six @cordisjs plugins) and every reference that resolves them: manifest names and dependency keys, module specifiers including declare-module merges, cordis.yml plugin names, tsconfig paths, every Markdown fence, and `docs/` prose. Directory names, upstream versions, and dependency ranges are unchanged, so vendor/README.md still reads as an upstream snapshot; its manifest table gains an upstream-name column so THIRD_PARTY_NOTICES keeps MIT attribution pointed at each fork's origin. The tutorial tier follows the rename end to end: its yaml fences named plugins the Loader can no longer resolve, its `ts ignore-check` fences disagreed with the compiled fences beside them, and its prose quoted both. The contracts that told readers to keep upstream names — the root convention and the vendoring cookbook's tree comment and manifest invariant — now say to rescope instead. Two rules read `@deepseek-ai/` as "another workspace plugin": the client bundle purity gate now names the vendored libraries a browser bundle inlines, and the files where a bare `cordis` is an agent-preset id keep that product data.
@deepseek-ai/dsh-storage-json
English | 中文
JSON backend for the storage hub: one human-readable <unit>.json file per unit under a configured root, registered as backend json. Design: domain KV storage Agent Note.
Model
- The in-memory unit state is authoritative; every write primitive republishes the whole file via temp-write + fsync + atomic
rename()replace. A unit file is always the complete current net state — legibility is this backend's reason to exist; scale is the SQLite backend's job. - A missing file opens as an empty unit and materializes on the first write. A foreign or unparsable file rejects with
malformed-medium; a stored version differing from the descriptor rejects withversion-mismatch(no migration, pre-release stance). - Write ordering across calls belongs to the caller (the domain layer's write chain); each single call is atomic and durable once resolved.
Config
| Key | Type | Default | Meaning |
|---|---|---|---|
root |
string | required — no default (a cwd fallback would scatter files) | Directory holding unit files; created 0o700 on demand |
Model Experience
Stored domain records
What the model sees
Nothing. This backend contributes no prompt, tool, or schema; it persists non-session domain data behind ctx.storage for host-side consumers only.
Token effect
Zero live-request tokens.
KV Cache effect
None — the backend never touches live request prefixes.
Known Limitations and Deferred Work
- Windows durability relies on libuv's
rename()(MoveFileExWwith replacement) without an explicit write-through flag; the session-log backend's stricter Win32 write-through publish helper is planned to move down here when the append-log facet lands (see the Agent Note's migration section). - No cross-process write locking: two processes writing the same root can interleave whole-file replacements (last write wins). Single-host-process deployments are the current consumer; the multi-process story is deferred per the Agent Note's out-of-scope table.