readme: tighten cost claim and agent copy

[abrahamwolke]

Summary

• "Save up to 40%" → "Save 40%+" — more direct and confident; "up to" undersells the claim
• "normal file-reading tools" → "native search & file reading tools" — more precise about the mechanism (search + read, not just read)

No code changes.

Summary by CodeRabbit

• Documentation
  • Updated README with revised performance claims
  • Clarified how agents utilize native search and file reading capabilities

[coderabbitai]

Walkthrough

The README receives two text refinements: the performance claim shifts from "up to 40%" to "40%+" for a stronger assertion, and the agent-compatibility explanation is clarified to specify that agents leverage their "native search & file reading tools" rather than generic "normal file-reading tools" when consuming .graph files.

Changes

Cohort / File(s)	Summary
README Documentation README.md	Strengthened performance claim from "up to 40%" to "40%+" and refined agent-compatibility language to reference agents' native tools instead of generic file-reading tools.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~2 minutes

Possibly related PRs

• Rewrite README: outcome-led headline, how it works, works with any agent #57: Overlaps with README updates regarding agent compatibility and how .graph files are consumed by agents—both PRs adjust wording around agent-native tooling and file-reading mechanisms.

Suggested reviewers

• jonathanpopham

Poem

📄 From "up to" we climb to "plus and more,"

Native tools now shine and soar,

Claims refined, just right and tight—

Small tweaks that make the docs feel bright! ✨

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 inconclusive)

Check name	Status	Explanation	Resolution
Description check	❓ Inconclusive	The description is concise but misses required template sections like 'Why', 'Test plan', and structured verification steps.	Add sections for motivation/context and test plan verification (even though no code changed, document testing approach for README review).

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately describes the main changes: updating README copy about cost claims and agent tool descriptions.
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 unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch readme/copy-updates

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@README.md`:
- Line 3: The headline "Save 40%+ on agent token costs with code graphs." is an
over‑general claim; update that copy (the exact string "Save 40%+ on agent token
costs with code graphs.") to qualify the savings—e.g., append "in our
benchmarks," "in tested scenarios," or "up to 40%+ in benchmarked workloads"
and/or add a parenthetical or footnote linking to the benchmark methodology so
readers don’t treat it as a universal guarantee.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 9384e4e7-9b7c-4c5a-ab58-b174aeef107d

📥 Commits

Reviewing files that changed from the base of the PR and between d2c910a and 9c5932e.

📒 Files selected for processing (1)

• README.md

[coderabbitai]

⚠️ Potential issue | 🟠 Major

Qualify the “40%+” claim to avoid over-promising.

“Save 40%+” sounds like a general guarantee. If this is benchmark-based, say that explicitly so readers don’t treat it as universal.

Suggested copy tweak

-Save 40%+ on agent token costs with code graphs.
+Reduce agent token costs by ~40% in our benchmarks with code graphs.

📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change

	Save 40%+ on agent token costs with code graphs.
	Reduce agent token costs by ~40% in our benchmarks with code graphs.

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@README.md` at line 3, The headline "Save 40%+ on agent token costs with code
graphs." is an over‑general claim; update that copy (the exact string "Save 40%+
on agent token costs with code graphs.") to qualify the savings—e.g., append "in
our benchmarks," "in tested scenarios," or "up to 40%+ in benchmarked workloads"
and/or add a parenthetical or footnote linking to the benchmark methodology so
readers don’t treat it as a universal guarantee.