1. Conduid
  2. RAG
  3. Code Meridian
MCP server · RAG

Code Meridian

Local graph memory for AI coding agents, giving MCP-compatible tools precise codebase context before they edit.

34Low

Scored 11 days ago · breakdown

About Code Meridian

Code Meridian is an MCP server in the RAG category: local graph memory for AI coding agents, giving MCP-compatible tools precise codebase context before they edit. It has been installed 0 times through Conduid.

Tools (40)

analyze_changed_subgraphSummarize changed-file risk, impacted nodes, tests, architecture smells, and docs from a bounded graph neighborhood

Takes no parameters.

analyze_feature_implementation_pathMap a feature request or docs/features file to implementation status, touched areas, tests, docs, missing graph evidence, and risk

Takes no parameters.

answer_change_context_challengeValidate the user's one or two selected answers, halt on mistakes, explain them, and allow retry

Takes no parameters.

build_minimal_contextBounded context pack with callers, callees, impact, tests, gaps, likely files, token estimate, and model guidance

Takes no parameters.

check_graph_freshnessReport graph confidence from indexed file, line, and timestamp metadata

Takes no parameters.

classify_keywordsClassify derived keywords as domain/technical/tooling/common/noise and persist usefulness scores

Takes no parameters.

clear_code_graphClear all indexed code graph nodes while preserving docs

Takes no parameters.

clear_project_knowledgeClear one project's indexed graph and docs before rebuilding

Takes no parameters.

execute_context_workflowExecute an approved read-only context workflow and return per-step results

Takes no parameters.

find_config_definitionsFind where a canonical config key is defined or overridden

Takes no parameters.

find_config_usageFind which code reads or binds a canonical config key

Takes no parameters.

find_connectionShortest path between two code elements

Takes no parameters.

find_coverage_gapsProduction code not called by tests

Takes no parameters.

find_duplicate_candidatesReview likely duplicate methods/classes with refactor-risk signals

Takes no parameters.

find_frontend_cascade_conflictsReport likely CSS/SCSS override conflicts from indexed specificity and source-order metadata

Takes no parameters.

find_graph_driftDetect stale graph data before relying on exact implementation targets

Takes no parameters.

find_impactBackward blast-radius analysis

Takes no parameters.

find_implementation_patternsFind reusable implementation shapes such as entry -> service -> repository -> tests

Takes no parameters.

find_implementation_surfaceRank likely files and symbols to edit for a feature goal

Takes no parameters.

find_related_knowledgeFind lexically related code and docs through shared keywords

Takes no parameters.

find_similar_nodesFind duplicate code patterns (requires embeddings enabled)

Takes no parameters.

find_smell_pathsShow shortest forbidden architectural dependency paths

Takes no parameters.

find_stale_knowledgeDetect stale docs, weak mentions, orphaned external concepts, and orphaned code references

Takes no parameters.

find_test_shieldMap direct test protection, indirect shields, and unshielded change-path nodes

Takes no parameters.

find_unreferencedDead-code candidates

Takes no parameters.

get_architectural_overviewProject map by namespace/module

Takes no parameters.

get_change_contextRetrieve opt-in human-cognitive-seed context with current hash/orphan status

Takes no parameters.

get_context_for_editingCompact callers/callees/interfaces context for a node

Takes no parameters.

hybrid_searchFind semantically similar code near a node or subsystem boundary

Takes no parameters.

knowledge_decayAlias of `find_stale_knowledge` for graph-native stale-knowledge review

Takes no parameters.

plan_context_workflowPlan the right sequence of CodeMeridian tools for an agent task

Takes no parameters.

query_codebaseNatural-language search over code structure

Takes no parameters.

rebuild_keyword_graphRebuild derived `Keyword` nodes and `HAS_KEYWORD` edges from indexed graph text

Takes no parameters.

record_change_contextPreserve one bounded attributed human-cognitive-seed decision or constraint against an exact code node

Takes no parameters.

record_change_context_challenge_noteSave an optional user-written note after the challenge is solved

Takes no parameters.

replace_surfaceGroup dependency replacement work into safe and risky clusters before a library migration

Takes no parameters.

resolve_exact_symbolResolve symbol/file/line hints to canonical node IDs before editing

Takes no parameters.

search_documentationSearch indexed README/ADR/documentation content

Takes no parameters.

start_change_context_challengeOpen an MCP App with 3-4 LLM-authored code choices grounded in an exact node and its change context

Takes no parameters.

suggest_extractionsRank tightly connected groups that look like good extraction candidates

Takes no parameters.

Install

Clone
git clone https://github.com/Driftya/code-meridian

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

Ask AI about Code Meridian

Powered by Claude · Grounded in docs

I know everything about Code Meridian. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

  • ·README presentNot checked yet.
  • ·License declaredNot checked yet.
  • ·Tests presentNot checked yet.
  • ·Dependencies pinnedNot checked yet.
  • ·No dynamic code executionNot checked yet.
  • ·Scoped permissionsNot checked yet.
Questions

About Code Meridian

How do I install Code Meridian?

Run git clone https://github.com/Driftya/code-meridian, then add the server to your MCP client's configuration. Conduid has recorded 0 installs, so the command is known to work with current clients.

Is Code Meridian safe to use with an AI agent?

Its trust score is 34 out of 100 (low). Conduid hasn't run static security checks on this repository yet, so review the source yourself before granting it credentials. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Code Meridian still maintained?

Conduid hasn't recorded a commit date for this repository yet. Check the repository directly for recent activity.