docs: make published metrics traceable

[CoreyLeath-code]

Summary

Makes published SentinelAI metrics traceable. The reproducible benchmark baseline now names its committed generator, exact command, and output file. Values with no committed generator or CI artifact remain visible but are explicitly classified as Design Targets (Not Measured).

No benchmark value was invented, regenerated, or changed.

Files changed

File	Why
README.md	Adds a source column for committed benchmark evidence, removes the unsupported static coverage badge, preserves unsupported historical audit/inventory values under Design Targets (Not Measured), and gives the PSI/KS threshold a compile/source command.
Metrics.md	Reclassifies the ungenerated C++/Go/LLM/container performance figures as design targets instead of measured benchmarks.

Metric evidence audit

Metric	Measured?	Evidence	Command	File
Timed evaluations	Yes	Seeded raw JSON	python benchmarks/run_benchmark.py --iterations 20000 --evaluation-samples 2000 --output benchmarks/latest.json	benchmarks/run_benchmark.py, benchmarks/latest.json
Mean latency	Yes	Seeded raw JSON	Same benchmark command	Same files
Median latency	Yes	Seeded raw JSON	Same benchmark command	Same files
P95 / P99 latency	Yes	Seeded raw JSON	Same benchmark command	Same files
Minimum / maximum latency	Yes	Seeded raw JSON	Same benchmark command	Same files
Throughput	Yes	Seeded raw JSON	Same benchmark command	Same files
Peak traced memory	Yes	Seeded raw JSON	Same benchmark command	Same files
Precision / recall / F1	Yes	Seeded raw JSON; synthetic labels	Same benchmark command	Same files
Confusion matrix	Yes	Seeded raw JSON; synthetic labels	Same benchmark command	Same files
Benchmark environment	Yes	Recorded in raw JSON	Same benchmark command	Same files
PSI / KS threshold	Yes—configuration, not a benchmark result	C++ decision rule	g++ -std=c++17 drift-engine/drift_engine.cpp -o drift-engine/drift_engine	drift-engine/drift_engine.cpp
Focused API tests (4 passed)	No	Historical audit only	None committed	README.md
Focused API coverage (24%)	No	Historical audit only	None committed	README.md
Tracked files	No	Historical inventory only	None committed	README.md
Python / Go / C++ / TypeScript file counts	No	Historical inventory only	None committed	README.md
Source NCLOC	No	Historical inventory only	None committed	README.md
Test file / declaration counts	No	Historical inventory only	None committed	README.md
Workflow / dependency / Dockerfile counts	No	Historical inventory only	None committed	README.md
Kubernetes / Helm / Terraform / monitoring counts	No	Historical inventory only	None committed	README.md
Docker Compose service URL count	No	Historical inventory only	None committed	README.md
C++ PSI time (< 2ms)	No	No committed generator or artifact	None	Metrics.md
Go P95 latency (180ms)	No	No committed generator or artifact	None	Metrics.md
LLM summary time (1.2s)	No	No committed generator or artifact	None	Metrics.md
Throughput (150 RPS)	No	No committed generator or artifact	None	Metrics.md
Container cold start (< 3s)	No	No committed generator or artifact	None	Metrics.md

A row named Compute Time does not exist in the current README or Metrics.md; no value was added.

Validation performed

• Verified benchmarks/run_benchmark.py produces the schema and fields stored in benchmarks/latest.json.
• Verified .github/workflows/benchmarks.yml executes the benchmark with 20,000 iterations and 2,000 synthetic evaluation samples, validates its schema/F1 floor, and retains the artifact for 30 days.
• Verified the PSI/KS limits are coded in drift-engine/drift_engine.cpp.
• CI is triggered on this PR. This change does not alter runtime or benchmark logic.

Risks

• Historical values move from apparent measurements to clearly labelled, non-release evidence. Consumers relying on those figures as current performance data must use the versioned benchmark instead.

Follow-up recommendations

• Add committed, CI-executed native C++, Go, LLM, container-startup, and end-to-end benchmark harnesses before promoting those design targets to measured claims.
• Add a committed inventory-generation script if codebase counts are intended to be release evidence.

[coderabbitai]

Warning

Review limit reached

@CoreyLeath-code, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 59 minutes

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: e84f7437-cb2a-4e89-bf2a-349379894ab3

📥 Commits

Reviewing files that changed from the base of the PR and between 3e1a234 and 5bd62a7.

📒 Files selected for processing (2)

• Metrics.md
• README.md

---

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.}

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.