resQ MCP Server
# ResQ PyPI Packages
[](https://github.com/resq-software/pypi/actions/workflows/ci.yml)
[](LICENSE)
Python packages for the [ResQ](https://github.com/resq-software) disaster response platform, published to PyPI under the [`resq-software`](https://pypi.org/org/resq-software/) organization.
## Packages
| Package | Description | Version |
|---------|-------------|---------|
| [`resq-mcp`](packages/resq-mcp/) | FastMCP server -- connects AI agents to drone fleet, simulations, and disaster intelligence | [](https://pypi.org/project/resq-mcp/) |
| [`resq-dsa`](packages/resq-dsa/) | Zero-dependency data structures & algorithms for search, rescue, and geospatial ops | [](https://pypi.org/project/resq-dsa/) |
## Architecture
```mermaid
graph TB
subgraph "resq-software/pypi"
subgraph "packages/resq-mcp"
MCP[resq-mcp<br/><i>FastMCP Server</i>]
DTSOP[DTSOP<br/>Digital Twin Simulations]
HCE[HCE<br/>Hybrid Coordination]
PDIE[PDIE<br/>Predictive Intelligence]
DRONE[Drone Fleet<br/>Telemetry & Control]
MCP --> DTSOP
MCP --> HCE
MCP --> PDIE
MCP --> DRONE
end
subgraph "packages/resq-dsa"
DSA[resq-dsa<br/><i>Zero-Dep DSA</i>]
BF[BloomFilter]
CMS[CountMinSketch]
GR[Graph + A*]
HP[BoundedHeap]
TR[Trie]
DSA --> BF
DSA --> CMS
DSA --> GR
DSA --> HP
DSA --> TR
end
end
AI[AI Clients<br/>Claude / VS Code / Cursor] -->|MCP protocol| MCP
APP[Python Applications] -->|pip install| DSA
```
## Quick Start
```bash
# Install a package
pip install resq-mcp # MCP server for AI agents
pip install resq-dsa # Data structures (zero dependencies)
```
## Development
```bash
# Clone and setup
git clone https://github.com/resq-software/pypi.git && cd pypi
./bootstrap.sh
# Work on a package
cd packages/resq-mcp && uv sync && uv run pytest
cd packages/resq-dsa && uv sync && uv run pytest
```
### Release Flow
```mermaid
graph LR
DISPATCH[Manual dispatch<br/><i>Actions tab</i>] --> SR[Semantic Release]
SR -->|feat: / fix:| BUMP[Version Bump + Changelog]
BUMP --> BUILD[Build sdist + wheel]
BUILD --> ATTEST[Sigstore Attestation]
ATTEST --> PYPI[Publish to PyPI]
PYPI --> DOCKER[Docker Image<br/><i>resq-mcp only</i>]
```
The two published packages use [python-semantic-release](https://python-semantic-release.readthedocs.io/) with [Trusted Publisher](https://docs.pypi.org/trusted-publishers/) OIDC. Conventional commits determine the version bump, but **releases are triggered manually** from the Actions tab — the branch ruleset rejects the bot's version commit, so the workflow is `workflow_dispatch` only.
## License
[Apache-2.0](LICENSE) -- Copyright 2025 ResQ Software
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose with no overlap: get_deployment_strategy generates optimized drone strategies, run_simulation triggers physics simulations, and validate_incident handles incident report validation. The descriptions clearly differentiate their roles in the disaster response workflow.
All three tools follow a consistent verb_noun naming pattern (get_deployment_strategy, run_simulation, validate_incident) with clear, descriptive names that accurately reflect their functions. There are no deviations in naming conventions.
With only 3 tools, the server feels somewhat thin for a disaster response domain that includes strategy generation, simulation, and incident validation. While each tool is valuable, additional tools for mission execution, status monitoring, or data retrieval would provide more complete coverage.
The tools cover key phases of disaster response (validation, strategy generation, simulation), but there are notable gaps in mission execution and monitoring. The get_deployment_strategy description mentions using update_mission_params to push to drones, but this tool is not included, creating a workflow dead end.