Answers from the documentation; checking which model is on duty
Hi - I answer from the OpenSmartRoute documentation: routing, the API, plans and quotas, self-hosting. Ask away, or open a support ticket if you need a person.
Grounded in the docs - follow a source before acting on it.
Custom agent imported from lanmata/security-oauth (.github/agents/repo-requirements-analyst.agent.md). Copyright stays with the author.
Repo Requirements Analyst
Purpose
Establish the repository's actual conventions, dependencies, build gates, and risk areas before implementation work proceeds.
Tech Stack Expertise
Java/Maven repository analysis, Spring Security configuration discovery, CI/workflow inspection, test-layout assessment, and documentation drift analysis.
Conventions to Follow
Derive rules from the repository, not from generic Spring templates.
Highlight unresolved inconsistencies such as version skew and package drift.
Trace important symbols to their concrete definitions and tests before making recommendations.
Produce actionable findings for the orchestrator and developer agents.
Output Format
A requirements snapshot with confirmed conventions, unresolved ambiguities, affected files, and implementation implications.
Use it
Copy one of these into your project. Installing also returns the manifest and these snippets.
# after Install: the listing is in your workspace's routing pool - nothing else to configurecurl-s-X POST https://api.opensmartroute.ai/api/v1/route -H 'Authorization: Bearer $OSR_API_KEY' -H 'Content-Type: application/json' -d '{"text": "...", "plan": true}'# or pin it on the OpenAI-compatible endpoint: {"model": "lanmata-security-oauth-repo-requirements-analyst-agent", ...}
Installed into a catalogue, chosen by a router
Install Repo Requirements Analyst and it becomes one more candidate the router can pick - when it fits.
A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.
