Stream the report to the terminal as the writer generates it #5

Merged

schaefera merged 5 commits from worktree-deep-research-streaming into master

2026-08-17 10:48:53 +00:00

claude-bot commented

2026-08-17 10:17:46 +00:00

Collaborator

Summary

Switches the writer phase from Agent::prompt to rig's stream_prompt, printing each text delta to stdout as it's generated instead of waiting for the full response
Also carries the already-in-progress starter.rs -> core.rs rename (CLI parsing/observability init consolidated into core, main.rs left as a thin entry point) and a schemars import fix in review.rs, since these were already applied uncommitted in the working tree this branched from

Test plan

cargo build -p deep_research
cargo clippy -p deep_research (clean)
Manual run against a local Ollama instance to confirm the report streams token-by-token on the terminal

## Summary - Switches the writer phase from `Agent::prompt` to rig's `stream_prompt`, printing each text delta to stdout as it's generated instead of waiting for the full response - Also carries the already-in-progress `starter.rs` -> `core.rs` rename (CLI parsing/observability init consolidated into `core`, `main.rs` left as a thin entry point) and a `schemars` import fix in `review.rs`, since these were already applied uncommitted in the working tree this branched from ## Test plan - [x] `cargo build -p deep_research` - [x] `cargo clippy -p deep_research` (clean) - [ ] Manual run against a local Ollama instance to confirm the report streams token-by-token on the terminal

claude-bot added 1 commit 2026-08-17 10:17:47 +00:00

feat: stream the report to the terminal as the writer generates it 55f4f7eeb3

Wires up starter.rs -> core.rs (CLI parsing and observability init moved
into core, main.rs left as a thin entry point) and switches the report
phase from Agent::prompt to rig's stream_prompt, printing each text
delta to stdout as it arrives instead of waiting for the full response.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

schaefera added 1 commit 2026-08-17 10:19:13 +00:00

simplify: use print! + flush for streamed chunks, not an explicit stdout lock 469b7cf8c6

print! already locks stdout per call, so the manual lock()/write!() was
extra ceremony over what the flush actually needed. Matches rig's own
cli_chatbot streaming example more closely.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

schaefera added 1 commit 2026-08-17 10:20:40 +00:00

perf: lock stdout once for the whole report stream, not per chunk 4f32dd83d6

print!/println! each acquire stdout's lock internally; doing that per
streamed chunk in a tight loop adds needless contention. Lock once up
front and write!/writeln! through the held handle instead — which also
means the trailing newline must go through that same handle rather than
println!, since re-locking from the same thread would deadlock.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

schaefera added 1 commit 2026-08-17 10:30:26 +00:00

simplify: match the streamed Text item without destructuring its fields b12f156a1b

No destructuring needed since only .text is used; matching the whole
Text struct and drops the now-unused rig::message::Text import.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

schaefera added 1 commit 2026-08-17 10:37:15 +00:00

refactor: extract the streamed-text draining loop into its own module f53fcbd937

write_text_stream() in the new stream.rs doesn't touch anything specific
to write_report (topic, findings, the agent) — it just drains a
MultiTurnStreamItem stream, writes each text chunk to a caller-provided
writer, and returns the accumulated string. Pulling it out lets it be
covered by unit tests against a mocked stream and an in-memory writer,
independent of a live model.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>