Local Coding AgentBLOCK
Local MCP coding agent for ChatGPT Web: run files/commands on your own machine via the OpenAI Secure Tunnel, with a Windows tray app and metrics dashboard.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Local Coding Agent
Turn your machine into a local MCP coding workspace for AI agents. Let an AI agent read files, edit code, run checks, inspect git, and show live health metrics from a local dashboard.
Works with
Compatible with any MCP client. Not affiliated with Anthropic, OpenAI, GitHub, or Microsoft.
The Community Edition code is open source under AGPL. The project name and brand assets are governed separately by the Trademark Policy.
2defb88435bdOBSERVED · 2026-10-08Connect
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. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add local-coding-agent --env AGENT_APPROVAL_TOKEN=${AGENT_APPROVAL_TOKEN} --env ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY} --env LCA_DESKTOP_BRIDGE_TOKEN=${LCA_DESKTOP_BRIDGE_TOKEN} --env LCA_LICENSE_PUBLIC_KEY_PEM=${LCA_LICENSE_PUBLIC_KEY_PEM} -- npx -y [email protected]{
"mcpServers": {
"local-coding-agent": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"AGENT_APPROVAL_TOKEN": "${AGENT_APPROVAL_TOKEN}",
"ANTHROPIC_API_KEY": "${ANTHROPIC_API_KEY}",
"LCA_DESKTOP_BRIDGE_TOKEN": "${LCA_DESKTOP_BRIDGE_TOKEN}",
"LCA_LICENSE_PUBLIC_KEY_PEM": "${LCA_LICENSE_PUBLIC_KEY_PEM}"
}
}
}
}Exposed tools (9)
7 read · 2 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
bug_fix | read | Investigates errors, points at relevant files, and proposes a focused fix plan. |
default | read | Private Local Codex Studio profile |
demo-skill | read | Temporary skill created by the regression test. |
docs_update | write | Updates bilingual Vietnamese/English docs with matching content. |
network_check | read | Runs network/customer diagnostics and writes a redacted report. |
release_prep | read | Prepares changelog, version notes, build checklist, and a release-readiness report. |
repo_setup | write | Helps install, verify, and diagnose Local Coding Agent setup problems. |
safety_review | read | Checks permission risks, token leaks, unsafe commands, public tunnel exposure, and log redaction. |
work | read | test |
Trust audit
BLOCKgrade F · trust 53/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (6 observation(s))
- Network
- declared (8 observation(s))
- Shell
- declared (6 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (25)
if (/\beval\s*\(/.test(added)) findings.push({ priority: "P1", loc, issue: "eval() usage — potential code injection" });findings.push({ priority: "P1", loc, issue: "exec() in Python context — verify input is sanitized" });console.log(`License token written to: ${resolve(options.out)}`);"payload": "eyJzY2hlbWFWZXJzaW9uIjoxLCJwcm9kdWN0IjoibG9jYWwtYWdlbnQtc3R1ZGlvIiwidmVyc2lvbiI6InY1LjAuMCIsImNyZWF0ZWRBdCI6IjIwMjYtMDgtMTNUMTQ6MTE6MzAuNzQyWiIsImZpbGVzIjpbeyJwYXRoIjoiY29yZS9hZ2VudC10b29s
const text = readFileSync(this.file, "utf8").replace(/^/, "");
text = text.replace(/^/, ""); // strip UTF-8 BOM (some Windows editors add it)
const secret = "sk-os-secret-never-plaintext";
const secret = "sk-test-secret-value-123";
const secret = "sk-proj-abcdefghijklmnopqrstuvwxyz123456";
'api_key: "hunter2hunter2hunter2"'
const secret = "sk-proj-DEADBEEF1234567890";
const text = redactText("Bearer abcdefghijklmnop\n-----BEGIN PRIVATE KEY-----\nsecret\n-----END PRIVATE KEY-----");const r = await call(client, "read_file", { path: "../../../../etc/passwd" });check("path traversal blocked", (await call("read_file", { path: "../../../etc/passwd" })).isError);await call("read_file", { path: "../../../etc/passwd" }, { expectError: true });const bad = makeLocalReportPath("/tmp/x", "../../etc/passwd", "log");for i in $(seq 1 30); do curl -fsS http://127.0.0.1:8799/healthz >/dev/null && break; sleep 1; done
TEST_ENDPOINT=http://127.0.0.1:8799/mcp AUDIT_LOG=data/audit.log npm run test:security:baseline
(`http://127.0.0.1:8790/ui`).
- Health: `curl http://127.0.0.1:8787/healthz` → expect `{"status":"ok",...}`.Open the dashboard at `http://127.0.0.1:8790/ui` to copy the same setup,
@types/react, @types/react-dom, @vitejs/plugin-react, electron, electron-builder, react, react-dom, typescript
@types/react, @types/react-dom, @vitejs/plugin-react, electron, electron-builder, react, react-dom, typescript
@modelcontextprotocol/sdk
@modelcontextprotocol/sdk
Gates applied: no_behavioural_pass.
2defb88435bdfull audit observations/trust-audit/mcp-server/longngn204__local-coding-agent.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 |
|---|---|---|---|---|---|
| 2026-10-08 | 2defb88435bd | BLOCK | F | 53 | first audit |
Questions
What is the Local Coding Agent MCP server?
Local MCP coding agent for ChatGPT Web: run files/commands on your own machine via the OpenAI Secure Tunnel, with a Windows tray app and metrics dashboard.
What tools does Local Coding Agent expose?
9 in total: 7 read-only, 2 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Local Coding Agent safe to connect to an agent?
No — not without reading the findings first. The audit graded it F (53/100) and found 2 critical or high issues in the source. Each one is listed on this page with the file and line it is on.
What credentials does Local Coding Agent need?
It reads AGENT_APPROVAL_TOKEN, ANTHROPIC_API_KEY, LCA_DESKTOP_BRIDGE_TOKEN, LCA_LICENSE_PUBLIC_KEY_PEM, LCA_RELEASE_PUBLIC_KEY_PEM, LCA_RELEASE_SIGNING_PRIVATE_KEY_FILE, LCA_UPDATE_PUBLIC_KEY_PEM, LCA_UPDATE_SIGNING_PRIVATE_KEY_FILE, MCP_AUTH_TOKEN and OPENAI_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Local Coding Agent run?
It speaks streamable-http, so it runs as a service you connect to over the network. It is published on npm as local-coding-agent at 5.0.1.
How current is this page?
The grade is for one exact copy of the source (2defb88435bd), read on 2026-10-08. The repository is watched and re-audited when it changes.