release: 0.18.1

[yuanhao]

Two commits since v0.18.0. Patch — no API change.

Fixed

Summarization retries panicked in debug builds. RetryConfig::delay_for_attempt documents a 1-indexed attempt and computes attempt - 1; llm_compaction.rs passed the raw 0..=max_retries loop variable, so the first retry underflowed usize. agent_loop.rs increments before calling and was correct — this was confined to compaction.

It landed on a detached task, so nothing surfaced: the summarization vanished, no briefing arrived, and compaction fell back deterministically. That is one of the behaviours #150 was filed about, so some of those fallbacks were this rather than summarizer speed.

A briefing rejected on fingerprint mismatch now reports what it cost. The event carried summary: None, so cost accounting off SummaryStats under-counted the most wasteful failure — contradicting the contract types.rs states explicitly.

Added

LlmCompaction says so when briefings keep losing the race. After five consecutive deterministic fallbacks it warns once, naming the likely cause; a splice that lands resets the streak and clears the latch. Gated on a request having actually been issued, so it cannot claim a cost in the inert configuration.

The module docs now lead with the summarizer choice, which is the actual cause: reusing the loop's ModelConfig is the obvious call and, for a slow loop model, the worst one.

Verification

Tests	627 passed, 0 failed
Clippy	clean under -Dwarnings
Feature combos	default / openapi / gasp green
Live price audit	26 compared, 0 drifted

Release commit is exactly two files — Cargo.lock is gitignored.

Still open: #150 — compact_headroom_turns defaults to Some(30), pinning the target ratio to its 0.15 floor for tool-heavy work. That is a defaults change for every user and wants measurement across growth rates first.

🤖 Generated with Claude Code