CodemcpSAFE
Coding assistant MCP for Claude Desktop
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
NOTICE: claude code is available with Anthropic's $20/mo subscription, so I consider codemcp fully obsolete. However, there are some good design ideas (especially around the Git-versioning scheme) that I eventually want to port into the current generation of agentic coding clis.
Make Claude Desktop a pair programming assistant by installing codemcp. With it, you can directly ask Claude to implement features, fix bugs and do refactors on a codebase on your computer; Claude will directly edit files and run tests. Say goodbye to copying code in and out of Claude's chat window!
codemcp offers similar functionality to other AI coding software (Claude Code, Cursor, Cline, Aider), but it occupies a unique point in the design space:
- It's intended to be used with Claude Pro, Anthropic's $20/mo
subscription offering. I like paying for my usage with a subscription plan because it means zero marginal cost for agent actions; no more feeling bad that you wasted five bucks on a changeset that doesn't work.
Note that if you have Claude Max ($100/mo), Claude Code can also be used with subscription based pricing. The value proposition for codemcp is murkier in this case (and it is definitely inferior to Claude Code in some respects), but you can still use codemcp with Claude Max if you prefer some of the other UI decisions it makes. (Also, it's open source, so you can change it if you don't like it, unlike Claude Code!)
- It's built around auto-accept by default. I want my agent to get as
far as it can without my supervision, so I can review everything in one go at the end. There are two key things that codemcp does differently than most coding agents: we forbid unrestricted shell, instead requiring you to predeclare commands the agent can use in `codemcp.toml`, and we Git version all LLM edits, so you can roll back
8473d2ff7c56OBSERVED · 2026-09-24Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add codemcp -- uvx codemcp
{
"mcpServers": {
"codemcp": {
"command": "uvx",
"args": [
"codemcp"
]
}
}
}Exposed tools (12)
8 read · 3 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
chmod | read | Changes file permissions using chmod. Unlike standard chmod, this tool only supports |
edit_file | write | This is a tool for editing files. For larger edits, use the WriteFile tool to overwrite files. |
glob | read | Fast file pattern matching tool that works with any codebase size |
grep | read | Searches for files containing a specified pattern (regular expression) using git grep. |
init_project | read | If and only if the user explicitly asks you to initialize codemcp with |
ls | read | Lists files and directories in a given path. The path parameter must be an absolute path, not a relative path. |
mv | read | Moves a file using git mv and commits the change. |
read_file | read | Reads a file from the local filesystem. The path parameter must be an absolute path, not a relative path. |
rm | destructive | Removes a file using git rm and commits the change. |
run_command | write | Run a command that is configured in codemcp.toml. |
think | read | Use the tool to think about something. It will not obtain new information or change the database, |
write_file | write | Write a file to the local filesystem. Overwrites the existing file if there is one. |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (6)
rm
.pre-commit-config.yaml
hash1 = hashlib.md5(s1.encode("utf-8")).hexdigest()hash2 = hashlib.md5(s2.encode("utf-8")).hexdigest()4. Configure claude-mcp with URL: ``http://127.0.0.1:8000/sse`` (replace the port if needed.)
curl -sSf https://astral.sh/uv/install.sh | sh
Gates applied: no_behavioural_pass.
8473d2ff7c56full audit observations/trust-audit/mcp-server/ezyang__codemcp.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-09-24 | 8473d2ff7c56 | SAFE | B | 89 | source changed, verdict held |
Questions
What is the Codemcp MCP server?
Coding assistant MCP for Claude Desktop
What tools does Codemcp expose?
12 in total: 8 read-only, 3 that write, and 1 that can delete or overwrite (rm). Every one is listed on this page with its risk.
Is Codemcp safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Codemcp need?
No credential environment variables were found in its source, so it appears to need none.
How current is this page?
The grade is for one exact copy of the source (8473d2ff7c56), read on 2026-09-24. The repository is watched and re-audited when it changes.