copilot-memory-store

Provides a local JSON-backed memory store for context engineering with GitHub Copilot and MCP clients. Includes a CLI REPL to add/search/compress/delete/purge/export memories, and an MCP stdio server exposing tools and resources to let Copilot agents read/write/search memories and inject shaped/compressed context for tasks (optionally using DeepSeek for summarization/context shaping).

Evaluated Apr 04, 2026 (17d ago)
Homepage ↗ Repo ↗ Ai Ml ai-ml mcp copilot context-engineering memory-store typescript cli vscode
⚙ Agent Friendliness
59
/ 100
Can an agent use this?
🔒 Security
19
/ 100
Is it safe for agents?
⚡ Reliability
31
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
78
Documentation
70
Error Messages
0
Auth Simplicity
95
Rate Limits
0

🔒 Security

TLS Enforcement
0
Auth Strength
10
Scope Granularity
0
Dep. Hygiene
45
Secret Handling
50

Local JSON storage with no documented auth/authorization boundaries increases risk if exposed to untrusted processes/users. DeepSeek API key is referenced via .env; README does not describe logging/redaction behavior. No TLS/RPC encryption is described because MCP is stdio (transport-local), so confidentiality depends on local host environment. Error handling, input validation boundaries, and path/file safety are not detailed in the provided README.

⚡ Reliability

Uptime/SLA
0
Version Stability
55
Breaking Changes
40
Error Recovery
30
AF Security Reliability

Best When

You want a lightweight, local-first MCP/CLI memory workflow for Copilot-style agents, and you can accept JSON-file storage and stdio MCP transport.

Avoid When

You need robust security controls for untrusted remote users or you require enterprise-grade compliance features and hardened multi-tenant access control.

Use Cases

  • Persisting developer preferences and architectural decisions as searchable memories
  • Retrieval-augmented context injection for Copilot tasks
  • Budget-constrained memory compression (deterministic and optional LLM-assisted)
  • Building or testing MCP-based agent workflows with tools like memory_write/search/compress/delete/export
  • Team knowledge capture: store decisions/tags/recency and recall them during refactors or implementations

Not For

  • Multi-user production deployments that require strong authentication/authorization boundaries
  • Use cases needing server-side storage, backups, or managed persistence
  • Environments where writing to a local file is unacceptable (e.g., strict sandboxing without filesystem access)
  • Applications requiring documented REST/HTTP API semantics, pagination, and rate limiting

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

OAuth: No Scopes: No

No explicit auth mechanism is documented for the MCP server/CLI. Configuration centers on local filesystem path (MEMORY_PATH) and optional DeepSeek credentials for LLM-assisted features.

Pricing

Free tier: No
Requires CC: No

Open-source MIT package; local execution implies no direct service pricing, but optional DeepSeek calls may incur third-party API charges.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • MCP prompts are not supported by VS Code GitHub Copilot (per README); prompts should be tested via MCP Inspector or other MCP clients.
  • The MCP server runs over stdio; agents must be launched with the correct command/args/build output (dist/mcp-server.js) and MEMORY_PATH environment.
  • Demo prefilled data may exist in project-memory.json; README advises deleting/overriding MEMORY_PATH before real use.

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for copilot-memory-store.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-04-04.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered