Test parser fuzz inference invariants

[TraderSamwise]

Summary

• add generated assistant prose that casually mentions Claude Code and OpenAI Codex
• assert omitted-tool parser mode matches explicit mode when transcript chrome is present
• add no-chrome fuzz cases proving casual tool mentions do not infer parser mode

Verification

• vitest run src/agent-output-parser-fuzz.test.ts
• eslint src/ --max-warnings=0
• tsc --noEmit
• yarn build
• vitest run

Summary by CodeRabbit

• Tests
  • Enhanced test coverage for the agent output parser with improved inference validation and additional test scenarios.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
app	Ready	Preview, Comment	Jun 6, 2026 10:38am

[coderabbitai]

Too many files changed? Review this PR in Change Stack to see how the pieces fit before you dive in.

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 3d71b52a-1220-47f2-a52a-3ef8387b666f

📥 Commits

Reviewing files that changed from the base of the PR and between b52fd74 and 4f6b29b.

📒 Files selected for processing (1)

• src/agent-output-parser-fuzz.test.ts

---

📝 Walkthrough

Walkthrough

This PR enhances test coverage for parseAgentOutput's tool inference behavior by adding a new fragment generator that produces assistant responses mentioning both "Claude Code" and "OpenAI Codex", wiring it into the fuzz test infrastructure, and verifying both that inference works correctly and that casual mentions alone do not trigger mode inference.

Changes

Tool mention inference testing

Layer / File(s)	Summary
Tool mention fragment and wiring src/agent-output-parser-fuzz.test.ts	New assistantToolMention fragment generator produces assistant responses with tool-specific markers and sentinel strings referencing both "Claude Code" and "OpenAI Codex". Fragment is added to both claude and codex entries in fragmentMakersForTool fuzz test mapping.
Inference and non-inference test coverage src/agent-output-parser-fuzz.test.ts	Main fuzz test now validates inferred parsing (parseAgentOutput(raw)) against explicit parsing (parseAgentOutput(raw, { tool })), checking that inferred.parser.tool matches expected tool and blocks are equal. New deterministic test confirms that assistantToolMention-only transcripts leave parser tool as "unknown" with a single response block.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~12 minutes

Possibly related PRs

• TraderSamwise/aimux#75: Adds the inference helper logic that these new fuzz and deterministic tests exercise for both success (tool is inferred) and boundary (only mentions, no inference) cases.
• TraderSamwise/aimux#68: Prior evolution of tool-mode fuzz coverage in the same test file, adding/adjusting transcript fragment makers and parsing assertions.
• TraderSamwise/aimux#71: Overlaps at the same fragmentMakersForTool wiring layer, extending tool-specific fragment lists in the fuzz test infrastructure.

Poem

🐰 A fragment for mentions, both Claude and Codex alike,
Now fuzzed and deterministic, the parser's inference in sight.
Inference works when we mean it, stays unknown when we're just casual,
Test coverage blessed—the tool detection's no longer accidental! 🔍

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'Test parser fuzz inference invariants' directly describes the main purpose of the changes: adding fuzz tests that verify parser inference behavior doesn't incorrectly infer tool mode from casual tool mentions.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch test/chat-parser-fuzz-expansion

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}