Skip to main content
Glama
CISO Assistant MCP Server

CISO Assistant MCP Server

README.md
# MCP server with the full medulla engine contract
uv pip install "ciso-assistant-api[mcp]"

# Everything (development)
uv pip install "ciso-assistant-api[all]"      # or: python -m pip install "ciso-assistant-api[all]"
```

### Container images (`:mcp`)

One multi-stage `docker/Dockerfile` builds the MCP-only image (`--target mcp`):

| Image tag | Build target | Contents | Entrypoint |
|-----------|--------------|----------|------------|
| `${CISO_ASSISTANT_MCP_IMAGE}` | `--target mcp` | `ciso-assistant-api[mcp]` — MCP + `medulla`, no agent runtime | `ciso-assistant-mcp` |

```bash
docker build --target mcp -t "${CISO_ASSISTANT_MCP_IMAGE}" docker/

TDQS

C2/5.0

Scored across 1567 tools

Disambiguation1/5

Within nearly every resource there are several indistinguishable tools: `*_retrieve`, `*_object_retrieve`, `*_retrieve_2`, `*_list`, plus literal duplicates like `quality_check_retrieve`/`quality_check_retrieve_2` and `graph_data_retrieve`/`graph_data_retrieve_2`. `applied_controls_*` and `policies_*` expose the same underlying model with near-identical descriptions, and the generic `ciso_assistant_api_request` tool overlaps the entire surface, so an agent cannot reliably choose a tool.

Naming Consistency4/5

Almost everything follows a predictable `ciso_assistant_api_<resource>_<action>` snake_case pattern, which is consistent at the string level. Minor deviations exist: `_retrieve_2`/`_object_retrieve` suffixes, the unprefixed `ciso_ingest`, and `ciso_assistant_api_request`, which break the pattern.

Tool Count1/5

1567 tools is an extreme mismatch for any agent-facing server; this is essentially a raw REST router dump rather than a curated tool set. Individual CRUD, sub-resource, export, and chart endpoints are all exposed separately with no grouping or pruning.

Completeness5/5

The surface covers virtually every CISO Assistant domain area — governance, risk, compliance, privacy, resilience/DORA, TPRM, evidence, metrology, chat, and serdes — with create/read/update/delete plus exports and analytics. If anything, it is over-complete rather than gapped.

Maintenance

ActivityNo data
ResponsivenessOutdated data