fixes by balureddy003 · Pull Request #27 · i-cuboid/WppAgentLayer

balureddy003 · 2026-03-02T02:34:39Z

Pull Request Template

⚠️ Before Submitting a PR, Please Review:

Please ensure that you have thoroughly read and understood the Contributing Docs before submitting your Pull Request.

⚠️ Documentation Updates Notice:

Kindly note that documentation updates are managed in this repository: librechat.ai

Summary

Please provide a brief summary of your changes and the related issue. Include any motivation and context that is relevant to your changes. If there are any dependencies necessary for your changes, please list them here.

Change Type

Please delete any irrelevant options.

Bug fix (non-breaking change which fixes an issue)
New feature (non-breaking change which adds functionality)
Breaking change (fix or feature that would cause existing functionality to not work as expected)
This change requires a documentation update
Translation update

Testing

Please describe your test process and include instructions so that we can reproduce your test. If there are any important variables for your testing configuration, list them here.

Test Configuration:

Checklist

Please delete any irrelevant options.

My code adheres to this project's style guidelines
I have performed a self-review of my own code
I have commented in any complex areas of my code
I have made pertinent documentation changes
My changes do not introduce new warnings
I have written tests demonstrating that my changes are effective or that my feature works
Local unit tests pass with my changes
Any changes dependent on mine have been merged and published in downstream modules.
A pull request for updating the documentation has been submitted.

github-actions · 2026-03-02T02:35:05Z

🚨 Unused NPM Packages Detected

The following unused dependencies were found:

📂 Root `package.json`

@opentelemetry/api
@opentelemetry/core
@opentelemetry/exporter-trace-otlp-http
@opentelemetry/sdk-node
oauth4webapi

📂 Client `client/package.json`

@codemirror/commands
@codemirror/lang-css
@codemirror/lang-html
@codemirror/lang-javascript
@codemirror/language
@codemirror/language-data
@codemirror/lint
@codemirror/search
@codemirror/state
@lezer/highlight
@react-hook/intersection-observer
@react-hook/window-size
@stitches/core
anser
clean-set
dequal
escape-carriage
lz-string
react-devtools-inline
react-is

📂 API `api/package.json`

@langchain/google-genai
@langchain/google-vertexai
@langchain/textsplitters
compressible
xml-crypto
xml-encryption
xml2js

⚠️ Please remove these unused dependencies to keep your project clean.

Copilot

Pull request overview

This PR tightens Woodland agent behavior by improving result ranking/ambiguity handling in the Product/Engine history tools, expanding routing/validation guardrails in the supervisor router, and refining tractor query normalization—along with config and test-result updates.

Changes:

Add/adjust scoring + “needs human review” gating for ambiguous/low-confidence matches in Woodland Product/Engine history tools.
Expand supervisor router keyword rules and add output validation guardrails for unsupported claims without citations.
Update Azure Search MCP config to use an env-provided index name and update related environment/test artifacts.

Reviewed changes

Copilot reviewed 10 out of 11 changed files in this pull request and generated 7 comments.

Show a summary per file

File	Description
tests/agents/woodland/results/accuracy_report.json	Updates recorded timing metrics for Woodland test runs.
librechat.yaml	Changes Azure Search MCP server index name from hardcoded to env-driven.
api/app/clients/tools/structured/WoodlandProductHistory.js	Adjusts attribute match scoring and adds additional “needs review” heuristics.
api/app/clients/tools/structured/WoodlandEngineHistory.js	Adds signal extraction + ranking/filtering and new “needs review” conditions.
api/app/clients/tools/structured/WoodlandAISearchTractor.js	Improves abbreviation/deck-size normalization and make/model extraction for multi-word brands.
api/app/clients/agents/Woodland/supervisorRouterAgent.js	Expands routing rules and adds validation warnings + guardrail behavior for unsupported/citationless claims.
api/app/clients/agents/Woodland/promptTemplates.js	Adds stricter instruction about not stating specifics without tool evidence.
api/app/clients/agents/Woodland/cyclopediaSupportAgent.js	Makes procedural safety import “optional” with a safe fallback.
api/app/clients/agents/Functions/initializeFunctionsAgent.js	Adds a test-mode fallback output path for tool invocation.
.env	Large env update including concrete secrets/credentials and formatting changes.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-02T02:41:01Z

+} catch (_) {
+  // leave safe fallback


The procedural safety classifier module exists in-repo (../../tools/structured/util/proceduralSafety), but this change silently swallows all require-time errors and falls back to treating everything as safe. If the module throws due to a regression, you’ll disable the technician-only escalation path without noticing. Catch only MODULE_NOT_FOUND (and log loudly when falling back), and rethrow other errors so safety checks can’t fail open.

Suggested change

} catch (_) {

// leave safe fallback

} catch (error) {

if (error && error.code === 'MODULE_NOT_FOUND') {

// eslint-disable-next-line no-console

console.warn('[CyclopediaSupportAgent] proceduralSafety module not found; using safe fallback classifier.', error);

} else {

throw error;

}

Copilot · 2026-03-02T02:41:02Z

 - If confidence is Low, state "needs human review" instead of guessing
 - Cite specific evidence for claims (SKU numbers, article titles, case IDs)
+- Do not state exact warranties, bundle composition/counts, pricing, blade sizes, or universal-fit claims unless directly returned by the selected domain tool response.
+- If a customer asks for those exact specifics and tool output is incomplete, ask one clarifying anchor question or escalate with "needs human review.".


There’s a punctuation/grammar error: the sentence ends with an extra period after the closing quote (..."needs human review.".). Remove the trailing period (and consider whether you want the period inside the quotes).

Suggested change

- If a customer asks for those exact specifics and tool output is incomplete, ask one clarifying anchor question or escalate with "needs human review.".

- If a customer asks for those exact specifics and tool output is incomplete, ask one clarifying anchor question or escalate with "needs human review".

Copilot · 2026-03-02T02:41:02Z

+      const requestedSignals = this._buildRequestedSignals(input);
+      const { ranked, activeSignalCount, topScore, ambiguous } = this._rankAndFilterResults(
+        results,
+        requestedSignals,
+      );
+
+      if (activeSignalCount > 0 && topScore <= 0) {
+        return this._serializeNeedsReview(
+          'NEEDS_HUMAN_REVIEW: No confident engine-history match for the provided model/engine cues.',
+        );
+      }
+
+      if (
+        !requestedSignals.isAttributeLookup &&
+        activeSignalCount >= 2 &&
+        ambiguous &&
+        topScore < 4
+      ) {
+        return this._serializeNeedsReview(
+          'NEEDS_HUMAN_REVIEW: Multiple close engine matches found; please confirm one deciding cue (engine label exact model, horsepower, or rake model/year).',
+        );
+      }
+
+      return this._serializeSuccess(ranked, top);


New result-ranking / ambiguity handling introduces additional NEEDS_HUMAN_REVIEW branches (e.g., topScore <= 0, ambiguous && topScore < 4). There are Jest prompt-scenario tests for EngineHistory, but none assert these new branches, so regressions here are likely to slip through. Add/extend tests to cover at least (1) zero-score with active signals, and (2) ambiguous close-match behavior.

Copilot · 2026-03-02T02:41:02Z

      AZURE_SEARCH_ENDPOINT: "${AZURE_AI_SEARCH_SERVICE_ENDPOINT}"
      AZURE_SEARCH_KEY: "${AZURE_AI_SEARCH_API_KEY}"
-      AZURE_SEARCH_INDEX_NAME: "wpp-knowledge-test-encyclopedia"
+      AZURE_SEARCH_INDEX_NAME: "${AZURE_AI_SEARCH_Encyclopedia_Index}"


The Azure Search MCP server now references ${AZURE_AI_SEARCH_Encyclopedia_Index}, but that env var does not exist in .env.example (and this repo already defines AZURE_AI_SEARCH_CYCLOPEDIA_INDEX, ..._WEBSITE_INDEX, etc.). This will resolve to an empty index name by default and break the MCP server at runtime. Use an existing variable name (e.g., AZURE_AI_SEARCH_CYCLOPEDIA_INDEX / a dedicated AZURE_AI_SEARCH_FAQ_INDEX) and add it to .env.example if it’s new.

Suggested change

AZURE_SEARCH_INDEX_NAME: "${AZURE_AI_SEARCH_Encyclopedia_Index}"

AZURE_SEARCH_INDEX_NAME: "${AZURE_AI_SEARCH_CYCLOPEDIA_INDEX}"

Copilot · 2026-03-02T02:41:03Z

+
+
+
+MONGO_URI=mongodb+srv://wppuser:3Zc7Vb1rtYUKFw2h@wpplibre.imudzax.mongodb.net/?retryWrites=true&w=majority&appName=wpplibre/LibreChat


This .env change includes a full MongoDB connection string with embedded username/password. Committing real credentials to the repository is a security incident (and makes rotating/handling secrets difficult). Replace with a placeholder value (as in .env.example), move real secrets to your deployment secret store, and rotate these credentials immediately.

Suggested change

MONGO_URI=mongodb+srv://wppuser:3Zc7Vb1rtYUKFw2h@wpplibre.imudzax.mongodb.net/?retryWrites=true&w=majority&appName=wpplibre/LibreChat

MONGO_URI=mongodb+srv://<username>:<password>@<cluster-hostname>/?retryWrites=true&w=majority&appName=<app-name>

Copilot · 2026-03-02T02:41:03Z

 CREDS_KEY=f34be427ebb29de8d88c107a71546019685ed8b241d8f2ed00c3df97ad2566f0
+
 CREDS_IV=e2341419ec3dd3d19b13a1a87fafcbfb


This commit adds encryption secrets (CREDS_KEY / CREDS_IV) directly into .env in the repo. These should not be checked in; keep them empty in tracked env templates and supply real values via environment/secret manager. Please rotate these keys since they are now exposed in git history.

Suggested change

CREDS_KEY=f34be427ebb29de8d88c107a71546019685ed8b241d8f2ed00c3df97ad2566f0

CREDS_IV=e2341419ec3dd3d19b13a1a87fafcbfb

CREDS_KEY=

CREDS_IV=

Copilot · 2026-03-02T02:41:03Z

 AZURE_AI_SEARCH_INDEX_NAME=wpp-knowledge-dev
-AZURE_AI_SEARCH_API_KEY=
+
+AZURE_AI_SEARCH_API_KEY=4n2O2WEXy5u0OdRv6DEsBotuSz7zzzm2iVspryzsKOAzSeAIFdGu


AZURE_AI_SEARCH_API_KEY is committed with a concrete key value. API keys must not be stored in the repo; keep this blank (like .env.example) and inject it via environment/secret manager. The key should be rotated after removing it from the PR, since it’s now exposed.

Suggested change

AZURE_AI_SEARCH_API_KEY=4n2O2WEXy5u0OdRv6DEsBotuSz7zzzm2iVspryzsKOAzSeAIFdGu

AZURE_AI_SEARCH_API_KEY=

fixes

17475e0

Copilot AI review requested due to automatic review settings March 2, 2026 02:34

balureddy003 merged commit 5d0dcda into main Mar 2, 2026
2 of 5 checks passed

Copilot started reviewing on behalf of balureddy003 March 2, 2026 02:35 View session

Copilot AI reviewed Mar 2, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fixes - #27

fixes#27
balureddy003 merged 1 commit into
mainfrom
upstream-sync

balureddy003 commented Mar 2, 2026

Uh oh!

Uh oh!

github-actions Bot commented Mar 2, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

-} catch (_) {
-  // leave safe fallback
+} catch (error) {
+  if (error && error.code === 'MODULE_NOT_FOUND') {
+    // eslint-disable-next-line no-console
+    console.warn('[CyclopediaSupportAgent] proceduralSafety module not found; using safe fallback classifier.', error);
+  } else {
+    throw error;
+  }

	- If a customer asks for those exact specifics and tool output is incomplete, ask one clarifying anchor question or escalate with "needs human review.".
	- If a customer asks for those exact specifics and tool output is incomplete, ask one clarifying anchor question or escalate with "needs human review".

	AZURE_SEARCH_INDEX_NAME: "${AZURE_AI_SEARCH_Encyclopedia_Index}"
	AZURE_SEARCH_INDEX_NAME: "${AZURE_AI_SEARCH_CYCLOPEDIA_INDEX}"




		MONGO_URI=mongodb+srv://wppuser:3Zc7Vb1rtYUKFw2h@wpplibre.imudzax.mongodb.net/?retryWrites=true&w=majority&appName=wpplibre/LibreChat

	MONGO_URI=mongodb+srv://wppuser:3Zc7Vb1rtYUKFw2h@wpplibre.imudzax.mongodb.net/?retryWrites=true&w=majority&appName=wpplibre/LibreChat
	MONGO_URI=mongodb+srv://<username>:<password>@<cluster-hostname>/?retryWrites=true&w=majority&appName=<app-name>

		CREDS_KEY=f34be427ebb29de8d88c107a71546019685ed8b241d8f2ed00c3df97ad2566f0

		CREDS_IV=e2341419ec3dd3d19b13a1a87fafcbfb

	AZURE_AI_SEARCH_API_KEY=4n2O2WEXy5u0OdRv6DEsBotuSz7zzzm2iVspryzsKOAzSeAIFdGu
	AZURE_AI_SEARCH_API_KEY=

Uh oh!

Conversation

balureddy003 commented Mar 2, 2026

Pull Request Template

Summary

Change Type

Testing

Test Configuration:

Checklist

Uh oh!

Uh oh!

github-actions Bot commented Mar 2, 2026

🚨 Unused NPM Packages Detected

📂 Root package.json

📂 Client client/package.json

📂 API api/package.json

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

📂 Root `package.json`

📂 Client `client/package.json`

📂 API `api/package.json`