Cascading review

By default every file in a PR gets the same deep review treatment. Cascading adds a triage step that classifies each file as skip, skim, or deep-review before the main review runs, so cheap files don’t consume the same token budget as critical ones.

Tiers

Tier	What happens
skip	File is excluded entirely — lock files, auto-generated code, vendored deps
skim	Lightweight single-pass review — diff-only context, no tools, simplified output schema (no suggestedFix, no ticket compliance)
deep-review	Full review pipeline — tree-sitter context expansion, code search tools, consensus voting, ticket compliance

Enabling it

Set a triage model to enable cascading automatically:

RUSTY_LLM_TRIAGE_MODEL=anthropic/claude-3-5-haiku-20241022

Or toggle explicitly:

RUSTY_CASCADE_ENABLED=true   # force on (requires RUSTY_LLM_TRIAGE_MODEL)
RUSTY_CASCADE_ENABLED=false  # force off even if triage model is set

How it works

1. The triage agent receives a truncated version of each file’s diff (≤200 tokens per file, 30k token budget total) and classifies it
2. Files that overflow the triage budget default to deep-review
3. Files the triage model misses also default to deep-review
4. Safety net: if triage classifies all files as skip, the top 20% by additions are force-promoted to deep-review
5. Skim-tier and deep-tier files are reviewed in parallel
6. Results from both tiers are merged, then passed through the judge (if enabled)
7. If triage fails entirely, the bot falls back to the standard full-review pipeline

Summary comment

When cascading is active, the PR comment includes a collapsible Triage Summary showing how many files were skipped, skimmed, and deep-reviewed, plus the triage model and token usage.

Cost

The triage call itself is cheap — truncated diffs and a small output schema. The savings come from skipping context expansion and tool calls for skim-tier files. For a typical PR where roughly 40% of files are config, docs, or tests, expect around 30–50% token reduction on the review calls.