kyle/boxmaker
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Brief P2: log and replay assistant messages exactly, drop the thinking-block risk

Browse Source 
Approved 2026-09-17. Evidence is in docs/decisions.md.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
kyle
2026-09-17 00:38:32 -07:00
co-authored by Claude Fable 5.1
parent 790136e94f
commit 8421d80e87
2 changed files with 6 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/decisions.md
+1 -1
View File
@@ -26,6 +26,7 @@ Approved by the owner on 2026-09-17. One commit each.
| # | Change | Evidence |
|---|---|---|
| P1 | Inference contract 6: the `tools` array is fixed per epoch. `find_tool` returns schemas as a tool result and the model calls them through a `call_tool(name, arguments)` meta-tool. Never instruct the model to call an undeclared tool. | M0 (h): adding a tool re-read the whole prompt; undeclared calls were coerced into `write_file`. |
| P2 | Inference contract 2: the session log stores assistant messages exactly as returned, including `reasoning_content`, and replays them unchanged. Remove the "known risk" about dropped thinking blocks. | M0 (e): Ornith's template keeps every think block. |
## Proposed changes to the design brief
@@ -33,7 +34,6 @@ From M0 and the kickoff review. None is applied yet. Each lands as its own commi
| # | Change | Evidence |
|---|---|---|
| P2 | Inference contract 2: the session log stores assistant messages exactly as returned, including `reasoning_content`, and replays them unchanged. Remove the "known risk" about dropped thinking blocks. | M0 (e): Ornith's template keeps every think block. |
| P3 | Inference contract 7: requests set `return_progress: true`; progress events count as liveness. | M0 (i): otherwise the stream is silent during prefill. |
| P4 | Inference contract 8: the thinking cap uses `reasoning_control` and the control endpoint. | README b10809. Not yet exercised. |
| P5 | Settle the open question: chat-completions with server-side tool parsing. | M0 (b), (c). |
docs/design.md
+5 -4
View File
@@ -84,7 +84,10 @@ ordinary outbound networking can reach the whole tailnet and the internet as the
measured with the server's tokenizer, enforced by a test.
2. **Append-only.** The request for turn N+1 is a strict extension of the request for turn N.
Nothing volatile (time, heartbeat notes, memory refreshes) goes anywhere but the newest message.
Enforced by a property test on the serialized message array.
Enforced by a property test on the serialized message array. The session log stores each
assistant message exactly as the server returned it (`content`, `reasoning_content`,
`tool_calls`) and replays it unchanged. Ornith's template keeps every thinking block, so
dropping or editing one changes the prefix.
3. **No in-place pruning.** Tool results are size-capped when first appended, never trimmed later.
4. **Compaction is an epoch change.** It happens only when the session is idle, writes a summary,
and starts a new epoch whose prefix is the baseline plus the summary. The old log is kept.
@@ -109,9 +112,7 @@ ordinary outbound networking can reach the whole tailnet and the internet as the
Open question settled by measurement in M0: use the server's chat-completions endpoint with
server-side tool parsing, or render the template in-process for byte-exact prefix control. Default
to chat-completions. Switch only if the M0 cache measurements fail. Known risk: Qwen-style
templates drop earlier thinking blocks from history, which changes the rendered prefix at the last
assistant turn.
to chat-completions. Switch only if the M0 cache measurements fail.
## Authority contract