SDLC Command Center MCP Server
Provides 18 tools for repository snapshot, release readiness, secret scanning, and approval-gated file writes with backup and audit log.
README
Autonomous SDLC Command Center
A local-first, safety-gated engineering control plane. It maps repositories, triages incidents, diagnoses CI failures, reviews security and reliability, scores delivery risk, measures release readiness — and can apply approval-gated file changes with backup, rollback, and a tamper-evident audit log.
What is in the box
| Layer | Contents | Platform |
|---|---|---|
| Skills | 7 workflow skills, each with a machine-readable contract.json and behavioral tests.md |
Any harness that loads Markdown skills |
| MCP server | mcp/sdlc_mcp_server.py — 18 tools over stdio or localhost HTTP |
Windows / Linux / macOS, Python 3.9+ |
| CLI | mcp/sdlc_cli.py — the full tool surface for any shell or CI runner |
Windows / Linux / macOS, Python 3.9+ |
| PowerShell commands | scripts/commands/*.ps1 — snapshot, readiness, preflight |
Windows PowerShell / pwsh, rg |
| Tests | scripts/tests/smoke.py (cross-platform, 39 tests) and scripts/tests/smoke.ps1 (20 tests) |
Python 3.9+ / pwsh |
| Lifecycle hooks | hooks/*.ps1 and hooks/*.sh |
pwsh / POSIX sh |
No third-party packages. No network calls against targets. No telemetry.
Quick start
# Cross-platform CLI (Windows, Linux, macOS)
python mcp/sdlc_cli.py doctor
python mcp/sdlc_cli.py snapshot --path /repo --include-git
python mcp/sdlc_cli.py tree --path /repo --max-depth 3
python mcp/sdlc_cli.py read-batch --path /repo --file README.md --file src/main.py
python mcp/sdlc_cli.py secret-scan --path /repo --format sarif
python mcp/sdlc_cli.py risk --path /repo
# MCP over stdio (any MCP host)
python mcp/sdlc_mcp_server.py
# MCP over localhost HTTP
python mcp/sdlc_mcp_server.py --http 8765 # POST /mcp, GET /health, GET /tools
# PowerShell commands (Windows)
.\scripts\commands\repo_snapshot.ps1 -Path <repository> -IncludeGit
.\scripts\commands\release_readiness.ps1 -Path <repository>
.\scripts\commands\plugin_preflight.ps1
.\scripts\tests\smoke.ps1
The gated write engine
Read tools are side-effect free. The three write tools (sdlc_write_file, sdlc_replace_in_file, sdlc_rollback) enforce safety in the engine, not in the prompt:
- Dry-run by default. Every call returns a unified-diff preview and changes nothing unless
confirm: trueis passed. - Confinement. Paths cannot escape the approved directory; traversal, symlinks, UNC/network paths,
.git, and the.sdlcstate directory are rejected. Sensitive names (.env, keys, credentials) needallowSensitive: true. - Backup first. Existing content is copied into
.sdlc/backups/<changeId>/before any mutation. Writes are atomic (temp file + rename). - Rollback.
sdlc_rollbackrestores originals (or deletes created files) bychangeId. - Audit. Every mutation appends to
.sdlc/audit.jsonl, a hash-chained log whose integrity is verified bysdlc_audit_log.
python mcp/sdlc_cli.py write --path /repo --file config.txt --content "v2" # dry-run diff
python mcp/sdlc_cli.py write --path /repo --file config.txt --content "v2" --confirm # apply (backup + audit)
python mcp/sdlc_cli.py changes --path /repo # list change sets
python mcp/sdlc_cli.py rollback --path /repo --change-id <changeId> --confirm # revert
python mcp/sdlc_cli.py audit --path /repo # verify the chain
Tool surface (18 MCP tools)
Read-only: sdlc_repo_snapshot, sdlc_release_readiness, sdlc_plugin_preflight, sdlc_read_file, sdlc_read_files, sdlc_directory_tree, sdlc_search_code, sdlc_secret_scan, sdlc_language_stats, sdlc_dependency_inventory, sdlc_git_history, sdlc_risk_score, sdlc_doctor, sdlc_list_changes, sdlc_audit_log.
Gated writes: sdlc_write_file, sdlc_replace_in_file, sdlc_rollback.
Rate limiting: The MCP server enforces a fixed-window per-tool rate limit (default 60 calls per 60 seconds per tool). Configurable via SDLC_RATE_LIMIT_CALLS and SDLC_RATE_LIMIT_WINDOW_SECONDS. Violations return as tool errors.
See Operating model, security notes, portability, and the MCP server guide.
Start with: Inspect this repository and produce a decision-ready SDLC control report.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.