Stop Tango garbage from polluting Opportunity Intelligence

[Occumed79]

What changed

Tango: precision-first secondary source

• Tango no longer runs an unbounded blank-query federal firehose when Fetch Intelligence has no search text; blank runs now use a focused occupational health upstream Tango search.
• Added a Tango-only fail-closed precision gate before the shared structured-opportunity judge.
• Generic clinical terms (vaccination, TB testing, audiograms, spirometry, drug testing, medical exams, etc.) now require employee/workforce, employment-screening, public-safety, deployment, or occupational/regulatory context when the source is Tango.
• Ordinary patient/community/student medical procurements and incidental medical boilerplate are rejected before they spend AI judge budget.
• Tango is now marked as a secondary/aggregator evidence source with medium confidence, rather than high-confidence primary evidence simply because the API itself is direct.

SAM.gov: make the single call useful

• Blank autonomous SAM runs still spend only one API request, but that request is now focused on the occupational health title instead of a broad unfiltered retrieval.
• This addresses the repeated zero-yield pattern caused by broad SAM results having thin metadata/description content that the local relevance model cannot reliably classify.

AI Discovery

• No tightening, throttling, or recall reduction. AI Opportunity Discovery is currently producing the useful results and remains unchanged.

Regression coverage

• Added Tango precision tests for true occupational-health scopes, workforce-context clinical components, patient/community false positives, and incidental-medical boilerplate.
• Updated SAM query-policy coverage for the focused one-call autonomous search.

Why

Production feedback showed most bad Opportunity Intelligence records were coming from Tango, SAM.gov was repeatedly returning zero usable records, and AI Discovery was producing the relevant opportunities. This PR fixes the two structured sources independently without degrading the useful AI discovery path.

[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.

[greptile-apps]

Your trial has ended. Reactivate Greptile to resume code reviews.

[coderabbitai]

Important

Review available on request

• 🔍 Trigger review

Reviews should be triggered manually for repositories with fewer than 10 stars. Select Trigger review above or comment @coderabbitai review to review the latest changes. For a full review, comment @coderabbitai full review.

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 522fd91d-cc5a-463c-a165-3615161a7ac2

---

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.}

[greptile-apps]

Your trial has ended. Reactivate Greptile to resume code reviews.

[greptile-apps]

Your trial has ended. Reactivate Greptile to resume code reviews.