codeTree

codetree is an MCP server that provides structured code understanding for coding agents using tree-sitter. It exposes 23 tools to extract file skeletons/symbols, inspect relationships (call graphs, references, blast radius), perform code quality analyses (complexity, dead code, clone detection), and support navigation/search and repo/graph queries (dependency graphs, change impact, dataflow/taint tracing, git history).

Evaluated Mar 30, 2026 (0d ago)
Homepage ↗ Repo ↗ DevTools mcp code-analysis tree-sitter static-analysis coding-agents developer-tools python graphs symbol-search
⚙ Agent Friendliness
65
/ 100
Can an agent use this?
🔒 Security
18
/ 100
Is it safe for agents?
⚡ Reliability
29
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
85
Documentation
80
Error Messages
0
Auth Simplicity
95
Rate Limits
0

🔒 Security

TLS Enforcement
0
Auth Strength
10
Scope Granularity
0
Dep. Hygiene
60
Secret Handling
30

Runs as a local MCP server over stdio; the README does not describe authentication, authorization, or transport security controls. It indexes local project files and caches to disk (.codetree/index.json, graph.db), so confidentiality depends on local environment protections. No explicit guidance is provided on avoiding logging of sensitive code contents or paths. Dependency hygiene is unknown; tree-sitter and fastmcp are third-party libraries.

⚡ Reliability

Uptime/SLA
0
Version Stability
40
Breaking Changes
40
Error Recovery
35
AF Security Reliability

Best When

You want a stateless, agent-friendly way to query code structure and relationships across multiple languages with minimal token usage.

Avoid When

You cannot allow local code indexing/SQLite caching, or you require strong guarantees about analysis correctness for complex semantic behaviors (e.g., advanced interprocedural effects) without further validation.

Use Cases

  • Help an AI coding agent understand a repository without reading entire files
  • Locate and retrieve specific symbols (classes/functions) and their source
  • Generate/inspect call graphs and reference relationships for impact analysis
  • Find dead code, duplicates/clones, and compute complexity hotspots
  • Support onboarding and documentation by finding undocumented functions
  • Perform lightweight static analysis workflows via an agent-friendly interface

Not For

  • Replacing a full compiler/toolchain type-checking and runtime testing
  • Security-critical vulnerability auditing (unless backed by robust, validated analysis and review)
  • Environments where running an indexing server over a large codebase is not acceptable
  • Use as a hosted SaaS API where strict uptime/SLA guarantees are required

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

OAuth: No Scopes: No

The README describes running an MCP server locally via `claude mcp add ...` and `uvx ... codetree --root .` with no mention of authentication or access control.

Pricing

Free tier: No
Requires CC: No

Self-hosted/open-source tool; no hosting pricing described.

Agent Metadata

Pagination
unknown
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • Indexing and graph persistence may require filesystem permissions and can take time on first run
  • If large repos are used, batch tools should be preferred to reduce repeated queries
  • Some analyses (e.g., dataflow/taint) may be approximate depending on language support and parser limitations
  • Tools that fetch full skeletons/symbols could still be token-heavy if used without the provided compact formats

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for codeTree.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-30.

6533
Packages Evaluated
19870
Need Evaluation
586
Need Re-evaluation
Community Powered