GitGuardianCAUTION
MCP server for remediating hardcoded secrets using GitGuardian’s API. It detects over 600 secret types and prevents credential leaks before code is made public.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Bring GitGuardian's secret detection and incident management into your AI agent. Scan code for credentials before they leak, triage existing incidents, generate honeytokens, and remediate findings — all from inside your IDE or chat client, backed by GitGuardian's 500+ detectors.
[!CAUTION] MCP servers are an emerging technology. Agents act on your behalf and under your responsibility. Use trusted MCP servers and review agent actions when they interact with tools. To limit blast radius the server defaults to read-only-leaning permissions; what is actually exposed is determined by the OAuth scopes your access token holds.
What it does
- Secret scanning — proactively scan files for leaked credentials.
- Incident management — list, filter, assign, resolve, and tag incidents
(both internal and Public Monitoring incidents).
- Honeytokens — generate honeytokens and list existing ones.
- Code-fix automation — open pull requests that remediate secrets in
repositories your workspace monitors.
The exact set of tools exposed to your agent depends on the OAuth scopes granted to your access token.
Prompt examples
Scan this codebase for any leaked secrets or credentials.
Remediate all incidents related to my project.
Check if there are any new security incidents assigned to me.
Help me understand this security incident and provide remediation steps.
List all my active honeytokens.
Generate a new honeytoken for monitoring AWS credential access.
Create a honeytoken named 'dev-database' and hide it in config files.
Quick start
The recommended way to run the GitGuardian MCP server is to point your MCP client at the hosted server. The MCP client handles OAuth automatically; no local install, no token to manage, no uvx.
Pick the URL that matches your GitGuardian region:
8fcc92f59108OBSERVED · 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 ggmcp -- uvx ggmcp==0.7.1
Exposed tools (16)
14 read · 2 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
failing_tool | read | raise RuntimeError( |
filtered_tool | read | return |
get_current_token_info | read | Return the current API token |
my_tool | read | pass |
needs_both | read | return |
needs_incidents | read | return |
ping_tool | read | return |
scan_secrets | read | Accept scan parameters without making an API request. |
test_tool | read | Test tool docstring. |
tool_no_scopes | read | return |
tool_with_incidents_read | read | Tool requiring incidents:read scope. |
tool_with_multiple | read | return |
tool_with_scan | read | Tool requiring scan scope. |
tool_with_teams_write | write | Tool requiring teams:write scope. |
tool_with_unavailable | read | return |
tool_with_write | write | return |
Trust audit
CAUTIONgrade B · trust 85/100 Install with care. The audit found things worth knowing before you trust its output.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | WARN |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- declared (8 observation(s))
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (17)
client = Client("http://127.0.0.1:8088/mcp", auth=os.getenv("GITGUARDIAN_PERSONAL_ACCESS_TOKEN"))os.environ["GITGUARDIAN_URL"] = os.environ.get("GITGUARDIAN_URL", "http://127.0.0.1:3000")GITGUARDIAN_URL=http://127.0.0.1:3000 \
.coderabbit.yml
.pre-commit-config.yaml
.release-please-manifest.json
.windsurfrules
module = importlib.import_module(dep)
importlib.import_module(package_name)
importlib.import_module(module)
"GITGUARDIAN_URL": f"http://127.0.0.1:{fake_gg_api.server_address[1]}","http://127.0.0.1:3000",
blablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablabl
blablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablabl
blablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablabl
blablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablabl
blablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablablabl
Gates applied: no_behavioural_pass.
8fcc92f59108full audit observations/trust-audit/mcp-server/gitguardian__gitguardian-1.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 | 8fcc92f59108 | CAUTION | B | 85 | first audit |
Questions
What is the GitGuardian MCP server?
MCP server for remediating hardcoded secrets using GitGuardian’s API. It detects over 600 secret types and prevents credential leaks before code is made public.
What tools does GitGuardian expose?
16 in total: 14 read-only, 2 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is GitGuardian safe to connect to an agent?
With care. The audit graded it B (85/100) and found 17 things worth knowing before you trust this server, listed below with the exact line each was found on.
What credentials does GitGuardian need?
It reads ENABLE_LOCAL_OAUTH, GITGUARDIAN_API_KEY, GITGUARDIAN_PERSONAL_ACCESS_TOKEN, GITGUARDIAN_TOKEN_LIFETIME and GITGUARDIAN_TOKEN_NAME 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 GitGuardian run?
It speaks stdio and streamable-http, so it runs as a local process your client starts. It is published on PyPI as ggmcp.
How current is this page?
The grade is for one exact copy of the source (8fcc92f59108), read on 2026-10-08. The repository is watched and re-audited when it changes.