Movie Recommendation MCP
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Movie Recommendation MCPrecommend me 3 sci-fi movies"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Movie Recommendation MCP
FastMCP wrapper around the Rust movie recommendation API.
This repository was extracted from
movie-reservation-platform-lab/movie-recommendation-service/axum-tools-mcp so
the MCP component can have an independent CI and artifact pipeline.
Run
uv run movie-recommendation-mcpThe production container runs as UID 10001 and listens on port 8092:
docker build --tag movie-recommendation-mcp:local .
docker run --rm --network host movie-recommendation-mcp:localThe optional development target keeps uv, curl, source, and test tooling in a
separate image without adding them to production:
docker build --target development --tag movie-recommendation-mcp:development .
docker run --rm --network host movie-recommendation-mcp:developmentThe production target starts again from the clean Python runtime stage and copies only the built virtual environment. Adding development tools therefore does not change the published runtime contents.
The demo assumes the MCP and recommendation API containers share one ECS task
network namespace, so the default downstream address remains
127.0.0.1:8082. Override the downstream URL when running the containers
independently.
Defaults:
MCP endpoint:
http://127.0.0.1:8092/mcpHealth endpoint:
http://127.0.0.1:8092/healthDownstream API:
http://127.0.0.1:8082
Useful environment variables:
MOVIE_RECOMMENDATION_API_URLAXUM_TOOLS_API_URLfor compatibility with the original demo extractionPORTHOSTOTEL_SERVICE_NAMEOTEL_EXPORTER_OTLP_ENDPOINTOTEL_EXPORTER_OTLP_PROTOCOLOTEL_RESOURCE_ATTRIBUTES
Related MCP server: Weather Info MCP Server
Tools
recommendation_get_moviesInputs:
limit, optionalpreference, optionalfault, optionaltraceparent, optionaltracestate, optionalcorrelation_id, optionalrequest_id, and optionaldemo_fault.Calls Rust
GET /recommendations.Success payload:
{"ok": true, "service_name": "movie-recommendation-mcp", "fault": "<fault-or-none>", "recommendations": [...]}Downstream error payload:
{"ok": false, "service_name": "movie-recommendation-mcp", "fault": "<fault-or-none>", "status_code": <int>, "error": "recommendation_dependency_failed"}
recommendation_healthInputs: optional
traceparent, optionaltracestate, optionalcorrelation_id, optionalrequest_id, and optionaldemo_fault.Calls Rust
GET /health.Success payload:
{"ok": true, "service_name": "movie-recommendation-mcp", "health": {...}}
Both tools forward:
traceparenttracestateX-Correlation-IdX-Request-IdX-Demo-Fault
Extraction Decisions
Keep the internal Python import path as
axum_tools_mcpfor this extraction slice. The public package, console script, service name, image, and repository are already namedmovie-recommendation-mcp; renaming the import path would add churn without changing the external MCP contract consumed by the agent.Keep the tool names
recommendation_get_moviesandrecommendation_health. The proven reservation-agent demo allowlistedrecommendation_get_moviesand called it withlimit,preference,fault, and propagation arguments.Keep
AXUM_TOOLS_API_URLas a compatibility fallback, but preferMOVIE_RECOMMENDATION_API_URLfor new runtime configuration.
Artifact Contract
Platform infrastructure consumes this component as an immutable application artifact, not as source code:
OCI image built from this repository's
Dockerfile.Private ECR image reference pinned by digest:
<account>.dkr.ecr.<region>.amazonaws.com/movie-recommendation-mcp@sha256:<64-hex-digest>.Separate release identifier, such as the Git commit SHA or release version, recorded alongside the digest by the consuming environment/infra repository.
Runtime command:
movie-recommendation-mcp.Container port and default
PORT:8092.MCP endpoint:
/mcp.Health endpoint:
/health.Required runtime configuration is environment-driven; the downstream API URL is supplied with
MOVIE_RECOMMENDATION_API_URLor the compatibility fallbackAXUM_TOOLS_API_URL.Deployment composition, environment manifests, promotion state, and AWS resources stay outside this repository.
Checks
uv sync --frozen
uv run --frozen --no-sync pytest tests
uv run --frozen --no-sync pytest automation/tests
uv run --frozen --no-sync ruff check .
uv run --frozen --no-sync ruff format --check .
uv run --frozen --no-sync python -m compileall src tests automationPushes to main publish a Linux AMD64 candidate to GHCR as
sha-<commit>-run-<run-id>-attempt-<attempt>. CI disables BuildKit's automatic registry attestation to keep
the candidate a single-image manifest, then records explicit GitHub build
provenance against the published digest for the environment admission gate.
Container security evidence
The pinned organization-owned actions publish the signed
recommendation-mcp-security-evidence-<run-id>-attempt-<attempt> artifact:
component-candidate-evidence-v1alpha3.json, verified image provenance,
CycloneDX SBOM, and subject-bound vulnerability report. Evidence is retained
for 14 days. Missing provenance or any CRITICAL finding without a current,
exact central exemption fails publication of the canonical evidence package.
Raw findings stay visible; covered findings produce passed-with-exemptions.
HIGH findings remain visible for admission review.
Run/attempt tags are discovery hints, not deployment selectors. Environment verification independently checks the successful canonical run and signed package before admitting its exact digest to ECR. This producer has no AWS credentials or deployment authority. Older runs without this package are not eligible for the new admission path; use a fresh successful main run. The environment reader must support v1alpha3 before admission. It independently reevaluates original findings against the latest approved central policy. See the shared action contract.
This canary adopts actions PR #18
at 036531133bcefd454b5afc0eb55f8ba0328901ea for both publisher actions and the
PR/local scanner. Prepare receives github-token: ${{ github.token }} for its
authenticated canonical-main lookup, using the publishing job's existing
contents: read permission. The job's other permissions remain required;
passing its token does not reduce that token's authority. This release also
hardens evidence failure paths (including bounded legacy report reads and safe
errors) and scanner cleanup. Evidence remains v1alpha3.
Offline caller tests verify wiring and guards. Hosted PR scanning does not
exercise prepare or prove private-repository access or canonical publication;
those require separate live rollout acceptance. Publication remains restricted
to push events on this repository's canonical main. Rollback reverts both action
pins, the PR tooling checkout, and the documented local tooling pin to
bb40579c285df0b581c48b10f9b34574d5c78639, and removes the new prepare token input
together. See the canary plan.
PR and local vulnerability checks
container-security-check builds the production linux/amd64 image on PRs and
other non-canonical runs. It uses the same reviewed shared tooling and v1alpha3
policy as publication, with only contents: read. The GitHub token is supplied
only to the scan/evaluation step to read approved policy from the actions repo.
An uncovered CRITICAL or a scanner/policy retrieval error fails the job.
The complete report, policy decisions and summary are retained for 14 days as
recommendation-mcp-pr-vulnerability-report-<run>-attempt-<attempt>, including
after a failed gate. These local-image diagnostics are not signed candidate
evidence. Canonical main publication independently scans its exact GHCR digest.
To reproduce using a sibling actions checkout at the reviewed commit:
git -C ../movie-platform-actions rev-parse HEAD
# Expected: 036531133bcefd454b5afc0eb55f8ba0328901ea
docker build --pull --platform linux/amd64 --target prod \
--tag movie-recommendation-mcp:local .
# Supply GH_TOKEN securely through your normal environment setup.
node ../movie-platform-actions/local-tools/container-security/lib/scan.mjs \
movie-recommendation-mcp:local \
--evidence-version v1alpha3 --component recommendation-mcpThe helper writes an ignored .local-container-security/run-*/ directory.
Exit 0 means policy pass, 1 means blocking findings, and 2 means an operational
or validation failure. Request exemptions separately in
the central governance repository
using the exact CVE/package/version/PURL from the complete report and supporting
applicability evidence. An unmerged request does not unblock the gate. After
maintainer approval reaches central main, re-run the PR check; no new producer
pin is needed for policy changes.
This server cannot be deployed
Maintenance
Related MCP Connectors
performance-review MCP — wraps StupidAPIs (requires X-API-Key)
jargon-translator MCP — wraps StupidAPIs (requires X-API-Key)
RAWG MCP — wraps the RAWG video games database API (rawg.io)
OMDb MCP — IMDB-derived movie / TV / episode data (BYO key)
Related MCP Servers
- AlicenseDqualityDmaintenanceAn MCP server that allows users to search for movies, get detailed information, receive genre-based recommendations, and discover popular/trending films using OMDb and TMDb APIs.59 npmMIT
- FlicenseNot gradedqualityDmaintenanceProvides weather information through MCP tools integrated with a FastAPI backend, enabling users to query current weather for single or multiple cities and check API health status.-
- FlicenseAqualityDmaintenanceAn MCP server that wraps The Movie Database (TMDB) API, enabling search for movies and TV shows, retrieval of movie details, recommendations, similar movies, trending content, streaming providers, and movie discovery.8-
- AlicenseAqualityBmaintenanceFastMCP server to interact with the Overseerr API for movie and TV show requests management, enabling users to request media, search, and manage libraries via natural language.81MIT