MCP, MCP server
repobrain
🧠RepoBrain (formerly Antigravity) — Give your repo a brain.
Compatibility: Not stated.
Grade C
One Medium finding. Read the findings before installing.
Not a guarantee. A hygiene check of the source at the commit shown, not a security audit.
Rescanned at commit 064e6a1ee1a3, unchanged since the check below.
INSTALL_REMOTE_SCRIPT
Why this grade
- INSTALL_REMOTE_SCRIPT Medium
Fetches a script from the network and runs it (a pipe to a shell, or an install hook with a URL).
docs/en/SANDBOX.md, line 31
curl -sSL https://get.microsandbox.dev | sh
Any High, or two or more Medium findings, give D. One Medium gives C. Low findings or a security Note give B. Each code names a finding behind the letter.
If you think this grade is wrong, email [email protected] with the entry link and your evidence. We aim to acknowledge within two working days and decide within seven. If we confirm an error, we aim to correct it within 24 hours of confirming it. These are aims, not guarantees.
Grades are a hygiene check of the source at the commit shown, on the date checked, not a security audit or a guarantee. The free directory grades are automated. Before release, a PlugVet engineer reviews every paid report and any grade we hold back, including all F grades, against the source code and the evidence we quote.
What was checked
Plain notes from the hygiene checks.
Permissions
Concern
can delete files.
PERMISSION_DELETE_FILES
benchmarks/codegraph-comparison/scripts/capture_trae_jsonl.py, line 39
completed = subprocess.run(
benchmarks/codegraph-comparison/scripts/judge_benchmark_v2.py, line 256
dirty = subprocess.run(
benchmarks/codegraph-comparison/scripts/judge_benchmark_v2.py, line 263
actual = subprocess.run(
Show more
benchmarks/codegraph-comparison/scripts/judge_benchmark_v2.py, line 278
completed = subprocess.run([
benchmarks/codegraph-comparison/scripts/judge_benchmark_v2.py, line 296
head = subprocess.run(
benchmarks/codegraph-comparison/scripts/prepare_corpora.py, line 40
completed = subprocess.run(
benchmarks/codegraph-comparison/scripts/prepare_corpora.py, line 86
completed = subprocess.run(
benchmarks/codegraph-comparison/scripts/prepare_corpora.py, line 123
shutil.rmtree(generated)
benchmarks/codegraph-comparison/scripts/prepare_corpora.py, line 160
shutil.rmtree(backup)
benchmarks/codegraph-comparison/scripts/prepare_corpora.py, line 175
shutil.rmtree(index)
benchmarks/codegraph-comparison/scripts/prepare_corpora.py, line 237
shutil.rmtree(index)
benchmarks/codegraph-comparison/scripts/run_benchmark_v2.py, line 121
completed = subprocess.run(
cli/src/rb_cli/cli.py, line 623
shutil.rmtree(item, ignore_errors=True)
engine/repobrain_engine/hub/storage.py, line 237
shutil.rmtree(path)
Install and update
Serious
install fetches a script from the network.
INSTALL_REMOTE_SCRIPT
docs/en/SANDBOX.md, line 31
curl -sSL https://get.microsandbox.dev | sh
docs/es/SANDBOX.md, line 32
curl -sSL https://get.microsandbox.dev | sh
docs/zh/SANDBOX.md, line 30
curl -sSL https://get.microsandbox.dev | sh
Network
Concern
can reach a host that is not written as a fixed address.
NETWORK_NO_FIXED_HOST
cli/src/rb_cli/cli.py, line 238
with urllib.request.urlopen(request, timeout=5):
cli/src/rb_cli/cli.py, line 244
response = requests.get(
engine/repobrain_engine/sandbox/microsandbox_exec.py, line 72
response = requests.post(
Show more
engine/repobrain_engine/tools/ollama_local.py, line 37
resp = requests.post(url, json=payload, timeout=60)
engine/repobrain_engine/tools/openai_proxy.py, line 60
response = requests.post(url, json=payload, headers=headers, timeout=30)
artifacts/benchmark-2026-05-09/grades_v4_tools.json, line 70
"note": "Correctly traces requests.get from the module helper through Session, PreparedRequest, adapter send, urllib3 urlopen, response wrapping, body read, and cleanup."
artifacts/benchmark-2026-05-09/questions_v4.json, line 14
{"id": "rq-x1", "type": "cross-cutting","q": "Trace what happens when a user calls `requests.get(url)` end-to-end: from module-level helper to actual socket I/O."}
Secrets
Concern
reads a credential or environment file.
SECRETS_ENV_READ SECRETS_PATH_DOTENV
cli/src/rb_cli/cli.py, line 141
"""Load simple KEY=VALUE entries from the workspace .env plus process env."""
cli/src/rb_cli/cli.py, line 142
env_path = workspace / ".env"
cli/src/rb_cli/cli.py, line 317
config = generation_config(Settings(_env_file=str(workspace / ".env")))
Show more
cli/src/rb_cli/cli.py, line 541
config_source = ".env" if env_exists else "process env"
cli/src/rb_cli/cli.py, line 553
reason = ".env missing" if not env_exists else f"missing {', '.join(missing)}"
cli/src/rb_cli/cli.py, line 554
_doctor_print(f"✗ .env config: {reason}; run rb-setup")
engine/install.sh, line 57
if [ ! -f ".env" ]; then
engine/install.sh, line 58
cat > .env << 'EOF'
engine/repobrain_engine/sandbox/microsandbox_exec.py, line 38
self._api_key = os.getenv("MSB_API_KEY", "").strip() or None
Agent instructions
Pass
No instruction to ignore earlier instructions or to hide a step.
Obfuscation
Note
a file was over the read cap and was not fully read.
OBFUSCATION_FILE_OVER_READ_CAP
benchmarks/codegraph-comparison/results/fair-e2e-turbo.json, line 1
{
Licence
Pass
A licence file is present (LICENSE).
Maintainer activity
Pass
The latest commit is within the last 365 days.
Note
Hygiene check of commit 064e6a1ee1a3. can delete files. install fetches a script from the network. can reach a host that is not written as a fixed address. reads a credential or environment file. a file was over the read cap and was not fully read.
Date and source
Date checked:
github.com
Commit graded: 064e6a1ee1a3364dbc34389c95bea5d8087f31e6
This link goes to the original host. PlugVet does not host or check what you download from there, and installing it is your own decision and risk.