| FMRS | 66 / 100 · C | 75 / 100 · B |
|---|
| Reliability | 10 / 20 | 12 / 20 |
|---|
| Security and permissions | 11 / 20 | 16 / 20 |
|---|
| Maintenance | 15 / 20 | 14 / 20 |
|---|
| Documentation | 16 / 20 | 18 / 20 |
|---|
| Setup experience | 14 / 20 | 15 / 20 |
|---|
| Best for | - AI agent builders who want local video editing, captioning, repurposing, and quality gating
- Developers and creators already using MCP clients such as Claude Code or Cursor
- Local media operators who want typed, guardrailed FFmpeg operations with traceable artifacts instead of guessed flags
| - Individuals or teams who prioritise privacy and offline operation and do not want data leaving the machine
- Users who share memory across multiple MCP clients such as Claude Code, Claude Desktop, Hermes Agent, OpenClaw, Cursor, Codex and Gemini CLI
- Scenarios that need encrypted storage, a hash-chained audit log and verifiable vault migration
- Users who want low-latency recall, a bundled embedding model and no extra LLM calls inside the memory layer
|
|---|
| Not for | - Users who want a hosted cloud editor or credit-metered SaaS
- Users who want an untyped shell wrapper for arbitrary FFmpeg commands
- Anyone unwilling to install FFmpeg (and optionally Node.js, Whisper, and other extras) locally
- People expecting to edit raw Insta360 .insv or GoPro .360 files directly — you must export a stitched equirect MP4 from the camera app first
| - Users who need cloud multi-device sync or a hosted team memory service; moving between machines means locking the vault and copying the file
- Users who expect the server itself to run an LLM that extracts facts and decides what to remember; Compartment explicitly keeps no LLM inside and leaves that to the host model
- Users who cannot keep a passphrase safe or who need a recovery path if it is lost, since Compartment never generates a password, seed or recovery phrase
- Deployments that require network transports such as SSE or streamable-http; this server is stdio only and opens no ports
|
|---|
| Required permissions | - Read and write local media files and output directories (editing, exporting, receipts)
- Execute local FFmpeg/ffprobe subprocesses
- Read local project/data directories (for example the ~/.mcp-video compatible data dir)
- Optional: network access for Hyperframes features, website capture, or installing Python extra dependencies
- Optional: C2PA signing requires c2patool plus a local manifest/signer
| - Read and write the vault file under the user's home directory, by default ~/.compartment/memory.vault, and its sibling settings file
- Access the session directory holding the unlock credential (configurable with COMPARTMENT_SESSION_DIR), which also carries the shared embedding process's Unix socket
- Optionally use the macOS keychain (compartment unlock --keychain is an explicit opt-in, and it survives reboots)
- Write or merge an mcpServers entry into each MCP client's own configuration file, taking a byte-exact backup first
- Install a PostToolUse hook for Claude Code that captures memory files the agent writes
- Serve a read-only dashboard on 127.0.0.1 behind a one-time random URL token
|
|---|
| Risks and side effects | - Agents may produce poor or broken video: the README explicitly requires video_quality_check, release checkpoints, and human visual/audio inspection before publishing
- Risk of overwriting source media: rescue and derivative flows claim the source stays immutable, but you should still back up important footage
- Optional extras are large (transcribe ~1 GB, stems ~2 GB, upscale ~2 GB), consuming significant disk and install time
- Depends on a local external toolchain (FFmpeg, plus Node.js 22+ for Hyperframes), so environment drift causes failures
- Optional capabilities such as website capture, local speech, or AI background removal may involve network access and extra licensing/compliance review
- Some README surfaces (desktop MCPB packaging, deeper Sonic World Audio integration, the trusted execution kernel, generative spend-cap plans) are gated, partial, or unreleased and should not be treated as stable shipped features
| - A leaked passphrase means the encrypted content can be decrypted; Compartment generates no recovery phrase and cannot recover a lost passphrase
- Enabling the memory_unlock tool places the passphrase in the model's context, which is why it is off by default
- If a 2FA keyfile is lost, for example a USB stick, the vault cannot be opened even with the passphrase
- `export --plaintext` writes the vault as unencrypted JSONL, so that file must be handled carefully
- Memory content can come from untrusted sources; recall wraps memories with a notice that they are stored data and content can be marked quarantined, but the host agent must still treat memory as data to limit prompt-injection risk
- The capture hook and client integration write into the user's own settings files, although the implementation backs up and merges rather than replacing
|
|---|
| Supported clients | Claude Code, Claude Desktop, Cursor | Claude Desktop, Claude Code, Hermes Agent, OpenClaw, Cursor, VS Code, Codex, Gemini CLI, Cline, Roo Code, Zed, OpenCode, LM Studio, AnythingLLM, BoltAI, Goose, Kiro |
|---|
| Tools | 19 | 14 |
|---|