Frontend Phase 13: ApiHost, route decorators, build fixes across 15 projects

[JustAGhosT]

Summary

• New ApiHost project with OpenAPI endpoint (/openapi/v1.json), CORS for localhost:3000, SignalR hub mapping, and controller discovery from 6 external assemblies
• Route decorators added to AdaptiveBalanceController and NISTComplianceController — [ApiController], [Route], HTTP method attributes, [ProducesResponseType], [FromBody]/[FromQuery]
• Build fixes across 15 projects — added FrameworkReference to 9 projects, resolved assembly name conflicts (ValueGeneration, AgencyRouter), fixed EF Core integration in AgentRegistry, rewrote CustomerIntelligence controller, added missing model types in DecisionSupport/ResearchAnalysis

Details

Area	Changes
ApiHost (new)	Web host, OpenAPI, CORS, SignalR, controller discovery
AdaptiveBalance	5 endpoints decorated (balance, override, history, learning-evidence, reflexion-status)
NISTCompliance	6 endpoints decorated (evidence, checklist, score, reviews, roadmap, audit-log)
AgentRegistry	EF Core fixes, nullable props, circuit breaker ctor, data models
AgencyRouter	Assembly name disambiguation, project reference fixes
ValueGeneration	Assembly name disambiguation, AuditEvent API migration
CustomerIntelligence	Controller rewrite with proper DI
DecisionSupport	Missing model types and stub components
ResearchAnalysis	Missing coordinator stub and XML docs
Directory.Packages.props	Added Microsoft.AspNetCore.OpenApi

Test plan

• dotnet build CognitiveMesh.sln — 0 warnings, 0 errors
• dotnet test CognitiveMesh.sln — 567 tests passing
• Verify OpenAPI spec serves at /openapi/v1.json when ApiHost runs
• Verify frontend can generate TypeScript client from spec (FE-001 next)

🤖 Generated with Claude Code

[greptile-apps]

Your free trial has ended. If you'd like to continue receiving code reviews, you can add a payment method here.

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: a645919a-5890-4194-8ea6-539b126621c8

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch feat/frontend-phase-13

---

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.}

[github-actions]

Summary

Summary

Generated on:	03/11/2026 - 00:23:49
Parser:	MultiReport (10x OpenCover)
Assemblies:	1
Classes:	4
Files:	3
Line coverage:	14.2% (3 of 21)
Covered lines:	3
Uncovered lines:	18
Coverable lines:	21
Total lines:	190
Branch coverage:	33.3% (2 of 6)
Covered branches:	2
Total branches:	6
Method coverage:	Feature is only available for sponsors

Coverage

CognitiveMesh.Shared - 14.2%

Name	Line	Branch
CognitiveMesh.Shared	14.2%	33.3%
CognitiveMesh.Shared.Interfaces.ChatMessage	0%	0%
CognitiveMesh.Shared.LogSanitizer	100%	100%
CognitiveMesh.Shared.Models.PolicyConfiguration	0%
CognitiveMesh.Shared.Models.RoutingRule	0%

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 5c2e73a30f

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[github-advanced-security]

CodeQL found more than 20 potential problems in the proposed changes. Check the Files changed tab for more details.

In general, to fix this kind of problem you should avoid relying on Object.ToString() for complex types in logs. Instead, either override ToString() on the type with a meaningful implementation, or explicitly convert the object to a useful string (for example, JSON serialization, or manual formatting of key properties).

Here, we should not modify AgentSearchCriteria because it is not shown, and we are constrained to editing only AgentRegistryService.cs. The safest and least invasive fix is to replace criteria.ToString() with a more informative representation localized to this logging call. Since we must not assume other project code, but can use well-known libraries or the BCL, a simple and dependency-free approach is to build a string from the known properties that are used in FindAgentsAsync: ActiveOnly, RequiredCapabilities, AgentTypes, MinimumAutonomyLevel, SearchText, and MaxResults. We then pass this formatted description into Sanitize(...) instead of the default ToString().

Concretely, within FindAgentsAsync’s catch block, change line 417 from:

Sanitize(criteria.ToString()));

to something like:

Sanitize(
    $"ActiveOnly={criteria.ActiveOnly}; " +
    $"RequiredCapabilities=[{string.Join(",", criteria.RequiredCapabilities ?? new List<string>())}]; " +
    $"AgentTypes=[{string.Join(",", criteria.AgentTypes ?? new List<string>())}]; " +
    $"MinimumAutonomyLevel={criteria.MinimumAutonomyLevel?.ToString() ?? "null"}; " +
    $"SearchText={criteria.SearchText ?? "null"}; " +
    $"MaxResults={criteria.MaxResults}"
));

This uses only standard .NET APIs (string.Join, null-coalescing) and stays entirely within the shown method, preserving existing behavior except that the log now contains detailed criteria values. It also continues to run the result through Sanitize(...) for safety.