mode-manager-mcp
mode-manager-mcp is a Python MCP server for VS Code/Copilot that manages persistent “memory” and “instruction” files (personal, workspace/team, and language-specific). It stores memories as Markdown files with YAML frontmatter and relies on VS Code Copilot chat integration to load the relevant memories on each turn.
Score Breakdown
⚙ Agent Friendliness
🔒 Security
Likely local/editor-integrated storage (Markdown + YAML frontmatter) with no described auth. This reduces network exposure but increases risk of sensitive data being written to workspace directories. README does not discuss encryption at rest, access controls, sanitization, or threat model for malicious prompt content. Dependency hygiene is unknown from provided data; manifest lists several third-party packages but no vulnerability review is provided.
⚡ Reliability
Best When
You want Copilot to have consistent context across conversations while you maintain memory as human-readable files inside your editor/workspace.
Avoid When
You cannot allow local workspace modifications, or you need centrally hosted/managed memory with enterprise-grade access controls and monitoring.
Use Cases
- • Persisting personal coding preferences and reminders for an AI assistant
- • Sharing team onboarding notes, coding conventions, and project best practices within a repo
- • Applying language-specific guidance automatically based on file types
- • Reducing repeated questions by reusing previously captured preferences and decisions
Not For
- • High-assurance or regulated environments requiring strong confidentiality controls beyond local file storage
- • Production systems needing audited APIs with formal contracts (OpenAPI/SDKs) and explicit operational guarantees
- • Workflows that cannot tolerate writing/updating instruction files in VS Code/Copilot directories
Interface
Authentication
No authentication mechanism is described in the README/manifest; it appears to run locally as an MCP server configured in VS Code.
Pricing
Open-source (MIT) library; cost is primarily local compute/editor usage rather than an external API/service.
Agent Metadata
Known Gotchas
- ⚠ Relies on VS Code/Copilot behavior for loading memories on each turn; agent-side testing may be sensitive to editor integration state.
- ⚠ Memory is stored as files; concurrency or repeated writes could lead to duplicates or conflicting updates if the tool logic is not explicitly designed for idempotent updates.
Alternatives
Full Evaluation Report
Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for mode-manager-mcp.
AI-powered analysis · PDF + markdown · Delivered within 30 minutes
Package Brief
Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.
Delivered within 10 minutes
Score Monitoring
Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.
Continuous monitoring
Scores are editorial opinions as of 2026-03-30.