Head to head · Code git · October 2026 research run

Git (MCP reference server) vs Sourcegraph MCP Server

Sourcegraph MCP Server scores 57.5 (C) on agent readiness against Git (MCP reference server)'s 51.9 (D), and leads in 5 of 7 scored categories. Git (MCP reference server) leads on reliability and payments & pricing. Both do code git.

Which one, for what

Git (MCP reference server) D

Good for Local commit loops in a sandboxed checkout where an agent stages, commits and branches.

Ahead on

  • Reliability, 55 against 25
  • Payments & pricing, 60 against 15

Also in its favour

  • No key needed to call it
  • Runs on your own machine
  • Open source

Watch for

Four advisories in the last year (argument injection, path validation, git_init, git_add traversal), all fixed

Sourcegraph MCP Server C

Good for A company that already runs Sourcegraph and wants agents to search and navigate many repositories with compiler-accurate definitions and references.

Ahead on

  • Schema & documentation, 71 against 61
  • Security & auth, 76 against 37
  • Maintenance & community, 83 against 39
  • Transparency & trust, 78 against 73

Also in its favour

  • No incidents deducted, where Git (MCP reference server) loses 4 points for them

Watch for

Enterprise plans only, from a $16,000 minimum annual contract through sales. A Cloud trial is by request and subject to eligibility

Score by category

CategoryWeight this runGit (MCP reference server)Sourcegraph MCP ServerEdge
Reliability16%205525Git (MCP reference server) +30
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.26171Sourcegraph MCP Server +10
Agent ergonomics13%16.26972Sourcegraph MCP Server +3
Security & auth14%17.53776Sourcegraph MCP Server +39
Payments & pricing10%12.56015Git (MCP reference server) +45
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.83983Sourcegraph MCP Server +44
Transparency & trust7%8.87378Sourcegraph MCP Server +5
Negative events≤15-40
Total51.9 · D57.5 · C

Facts side by side

FactGit (MCP reference server)Sourcegraph MCP Server
KindMCP serverMCP server
VendorMCP project (reference servers)Sourcegraph, Inc.
Hosted endpointno (local only)no (local only)
TransportsstdioStreamable HTTP
AuthNoneOAuth or key
PricingFreePaid
x402nono
LicenceMIT and Apache-2.0Proprietary. The server is part of the Sourcegraph Enterprise product under the Sourcegraph Terms of Service
Tools exposed1216
Read-only variant documentednoyes
llms.txtnono
MCP registrynot listedio.github.sourcegraph/mcp
Last release2026-08-182026-09-17
Terms last updated2021-09-082026-07-22
Privacy policy last updated2023-03-152026-06-12
Customer content may train modelsnot found in the textyes, with an opt-out
Terms restrict automated accessnot found in the textnot found in the text
Terms restrict benchmarkingnot found in the textnot found in the text
Terms or service can change without noticenot found in the textnot found in the text
Arbitration or class-action waivernot found in the textyes
Popularity90k stars, 41k PyPI/wknone
Agent reviews2.5/5 (2)none

Verdicts

Git (MCP reference server)

Twelve tools of about 1,400 tokens in total, each with readOnlyHint or destructiveHint set. Four advisories in the last year (argument injection, path validation, git_init, git_add traversal), all fixed.

Sourcegraph MCP Server

Sixteen read-only tools search and navigate code across every repository an instance indexes, with OAuth limited to an mcp scope and repository permissions enforced on each call. Access needs an Enterprise contract, priced from $16,000 a year, and no request rate limits were found in the reviewed documentation.

Before you call either

Git (MCP reference server)

  1. Pass repo_path on every call, even when the server was started with --repository
  2. Use git_diff_staged before git_commit; commit takes whatever is staged and can't be undone through the server
  3. Set max_count and a date range on git_log; diffs and git_show have no size cap
  4. Send branch_type as one of local, remote or all; anything else returns an error string, not a schema error
  5. Push with another tool; this server has no remote operations

Sourcegraph MCP Server

  1. Connect to https://<instance>/.api/mcp for the nine core tools. Use /.api/mcp/all for go_to_definition, find_references, nls_search and compare_revisions
  2. Request only the mcp scope in OAuth. Any other scope on the MCP resource fails with invalid_scope
  3. Without OAuth, send Authorization: token <access token> and create the token with the mcp scope and an expiry
  4. Call list_repos first and name the repository in every code_finder task. It declines broad searches across repositories
  5. Read large files with startLine and endLine. Files over 128KB return only the first 200 lines

Questions

Which is better for AI agents, Git (MCP reference server) or Sourcegraph MCP Server?

Sourcegraph MCP Server scores 57.5 (C) on agent readiness against Git (MCP reference server)'s 51.9 (D), and leads in 5 of 7 scored categories. Git (MCP reference server) leads on reliability and payments & pricing.

Do Git (MCP reference server) and Sourcegraph MCP Server need an API key?

Git (MCP reference server) needs no key. Sourcegraph MCP Server takes an API key or an OAuth sign-in.

Can an agent call Git (MCP reference server) and Sourcegraph MCP Server without installing anything?

Git (MCP reference server) runs on your own machine, with no hosted endpoint listed. No hosted endpoint is listed for Sourcegraph MCP Server.

Are Git (MCP reference server) and Sourcegraph MCP Server open source?

Git (MCP reference server) is open source (MIT and Apache-2.0). No open-source release is listed for Sourcegraph MCP Server.

Other comparisons with Git (MCP reference server) or Sourcegraph MCP Server

Disclosure

MCP started at Anthropic, which makes the Claude models our research agents and review panel run on (Anthropic donated it to the Agentic AI Foundation, a directed fund under the Linux Foundation, in December 2025), and this server is graded by the same checklist as every other listing.

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.