Tianyi Cui
2bae18f811
The loop logs the assistant/message (carrying tool-call blocks) BEFORE running the tools, so a crash mid-tool leaves durable tool calls with no matching tool/result. interruptedTurnClosers only added step/end + turn/end, so a resumed session's deriveMessages() replayed a dangling assistant tool-call — which every provider rejects as an invalid transcript on the next request. interruptedTurnClosers now scans the interrupted turn for tool-call blocks without a matching tool/result and synthesizes an error tool/result for each (before the step/end), so the rehydrated history is a valid transcript. Adds a dedicated repair.spec.ts and a shared-contract case proving both backends pair every orphaned call with a result. Docs (ADR 0018, both persistence READMEs, load() JSDoc) updated. Also fixes the echo-agent README session-cleanup path: demo:echo runs from the repo root, so sessions land in <repo-root>/.sessions/_no-cwd/, not examples/echo-agent/.sessions/ (review #33).
Architecture Decision Records
Short, immutable records of the why behind decisions that shape this codebase. Code and docs say what the system does; ADRs say why it does it that way and what we gave up.
Format: one file per decision, numbered, with Status / Context / Decision / Consequences. An ADR is never edited into a different decision — supersede it with a new one and cross-link.
When to write an ADR
Write one when a decision is all three of: durable (it shapes the codebase beyond a single function or package), contested (there was a real alternative you rejected, and a reasonable engineer might have chosen it), and surprising (a future reader would otherwise ask "why on earth is it done this way?"). The ADR captures the why and what we gave up — the parts code and docs can't.
Do NOT write an ADR for: a mechanical or local choice (a variable name, a one-file refactor); anything already enforced and explained by a gate or a convention in AGENTS.md; or a still-provisional decision tagged TODO(...) in the code — record those as TODOs and promote to an ADR only once they settle. When in doubt, the test is the "why on earth" question: if the code alone would mislead a careful reader about intent, write the ADR.
| # | Title | Status |
|---|---|---|
| 0001 | Vendor Cordis as source, not npm dependencies | accepted |
| 0002 | Microkernel: extension via Cordis event taxonomy, one concrete loop | accepted |
| 0003 | Event-sourced sessions with derived message history | accepted |
| 0004 | Provider-neutral content-block vocabulary owned by dsh-llm | accepted |
| 0005 | Custom typed tool-schema DSL instead of schemastery | accepted |
| 0006 | Tool schemas are part of the system-prompt assembly | accepted |
| 0007 | Mechanical quality gates over prose guidelines | accepted |
| 0008 | tsdown for JS bundling instead of dumble | accepted |
| 0009 | Capability seams — interface / implementation / consumer split | accepted |
| 0010 | Two LLM adapters as a design-verification twin | accepted |
| 0011 | Runtime arg validation at the model boundary | accepted |
| 0012 | Dev-mode invariants over compile-time deep-readonly | accepted |
| 0013 | Property-based testing for protocol-shaped code | accepted |
| 0014 | Doc-sync enforcement (doc code blocks + event taxonomy) | accepted |
| 0015 | Structured error taxonomy (HarnessError base) | accepted |
| 0016 | pnpm as the package manager instead of Yarn 4 | accepted |
| 0017 | Every session event is enclosed in a turn | accepted |
| 0018 | Session persistence as an abstract service over SessionEvent |
accepted |