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

Merge remote-tracking branch 'origin/codex/goal-session' into codex/commands

Browse Source 
# Conflicts:
#	packages/examples/stdio-demo/README.md
#	packages/ui/tui/README.md
This commit is contained in:
Tianyi Cui
2026-07-19 22:40:13 +08:00
parent 2850c22b7a 34fe167775
commit 4cb9c7ada2
255 changed files with 1757 additions and 507 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
packages/examples/stdio-demo/README.md
View File

@@ -80,15 +80,31 @@ Swap `llm-deepseek` for a `mock-llm` leaf plugin and you have the echo demo —
### Composed terminal agent request
**What the model sees**: Through `dsh-agent-spine-demo`, the `main` agent receives the harness identity, configured persona, skill catalog, and visible tools; this app also composes the generated [`ask_user_question` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-ask-user). Each ordinary terminal submission becomes a user message; submissions made while the agent runs steer the active turn. TUI commands remain outside model context.
#### What the model sees
**Token effect**: Child prompt and schema costs repeat per request; user input and tool history grow until compaction. Terminal banners, logger output, cards, and rendered transcripts add zero model tokens.
Through `dsh-agent-spine-demo`, the `main` agent receives the harness identity, configured persona, skill catalog, and visible tools; this app also composes the generated [`ask_user_question` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-ask-user). Each ordinary terminal submission becomes a user message; submissions made while the agent runs steer the active turn. TUI commands and their results remain outside model context.
#### Token effect
Child prompt and schema costs repeat per request; user input and tool history grow until compaction. Terminal banners, logger output, cards, and rendered transcripts add zero model tokens.
#### KV Cache effect
User and tool history is append-only while the composed prompt, schemas, child model route, and session prefix remain fixed. A composition change or compaction may invalidate reuse from its first changed token; terminal rendering has no cache effect.
### Human-answer result
**What the model sees**: Through `dsh-tool-ask-user`, successful terminal answers use that package's exact compact JSON shape. Interruption becomes exactly `Error: ask_user_question was interrupted before the user answered`; a closed stdin becomes `Error: ask_user_question cannot be answered because stdin is closed`.
#### What the model sees
**Token effect**: Only a completed or failed tool call adds retained result tokens; prompts printed while waiting are terminal-only.
Through `dsh-tool-ask-user`, successful terminal answers use that package's exact compact JSON shape. Interruption becomes exactly `Error: ask_user_question was interrupted before the user answered`; a closed stdin becomes `Error: ask_user_question cannot be answered because stdin is closed`.
#### Token effect
Only a completed or failed tool call adds retained result tokens; prompts printed while waiting are terminal-only.
#### KV Cache effect
Append-only; newly visible content follows the reusable request prefix and does not invalidate existing KV-cache entries.
## Known Limitations and Deferred Work