Handover
About
Shared, versioned continuation context for people and AI agents, with MCP publishing, review, annotations, identity, and organization isolation.
Details
- Author
- Unknown
- Categories
- AI
Jump to
Connect an agent and read the handover back
Start with browser sign-in and one complete round trip. Handover attributes the agent's work to your user while preserving your company, workspace, and access boundaries.
Done means read-back, not just connection.The identity is correct, the handover exists, and a later authorized actor can retrieve the durable record without the original chat.
Create a scoped credential in Company > Agents and store it as HANDOVER_TOKEN. The agent identity, scope, and revocation status remain separate from any employee account.
Interactive hosts use browser sign-in by default. Cline and unattended runners can use a named service credential when they do not support the standard OAuth flow.
Step 1codex mcp add handover --url https://handover.sh/api/mcp?profile=core
Add Handover at user scope. Claude Code discovers the authorization server and opens the browser sign-in automatically when it connects.
Configurationclaude mcp add --transport http --scope user handover https://handover.sh/api/mcp?profile=core
Add the canonical Streamable HTTP endpoint at user scope. Gemini can use Handover's browser authorization flow instead of a pasted employee token.
Configurationgemini mcp add --transport http --scope user handover https://handover.sh/api/mcp?profile=core
Configuration{ "mcpServers": { "handover": { "url": "https://handover.sh/api/mcp?profile=core" } } }
Open Cline's MCP wizard, choose Remote (HTTP), then add the separately named service credential through its private Static headers prompt.
Configurationcline mcp install handover --transport http https://handover.sh/api/mcp?profile=core
Install the dependency-free command from npm with Node.js 22.13 or newer. The package is MIT licensed and built from the public Handover MCP repository.
Install from npmnpm install --global handover-sh
Direct installercurl -fsSL https://handover.sh/install.sh | sh
Review thepublic package source,install.sh, andhandover.mjs. Set HANDOVER_URL and HANDOVER_TOKEN for the CLI or another unattended service identity; browser-connected MCP hosts do not need it.
Skills teach compatible agents how to publish, resume, review, and govern work safely. MCP or the CLI supplies the authenticated capability; the skill supplies the dependable workflow.
- Runhandover doctorfor a read-only CLI diagnostic.
- Search for a known handover.
- Read its current revision and one artifact.
- List open annotations before writing.
- Continue with expectedRevisionId.Download the complete workflow
The host lists Handover's tools,handover.whoamireturns the intended person or service agent, and search results are limited to context that identity may access.
- The MCP transport connects without a JSON or sign-in error.
- The identity belongs to the expected personal or company workspace.
- Reads and writes preserve that identity as the author.
- Revoking the credential immediately prevents further access.
Local-first agent memory: a plain-Markdown Obsidian vault is the source of truth, with a rebuildable DuckDB index for hybrid BM25 + vector + graph recall.
Persistent memory and semantic search for AI coding assistants across sessions
Give your agent a memory: shared, cited, tenant-isolated knowledge-graph memory for any MCP host. Grounded answers from a local-first June endpoint — abstains rather than guesses.
Decentralized persistent memory for AI agents — encrypted vault storage built on Walrus and Sui.
Persistent memory for AI assistants and coding agents across ChatGPT, Claude, Cursor, and other MCP-compatible tools.
Your portable AI memory vault — memories, skills & configs, shared across every AI tool.
Local Work Model for AI agents that learns from real outcomes.
Adaptive MCP memory system for AI applications. Learns which retrieval strategies work for your data, scores results using cognitive science models, builds a knowledge graph automatically, and validates every parameter change against real query history before adopting it. Patent pending.
Auditable, self-improving knowledge & memory for AI agents over MCP — citation-enforced answers and a replayable why-trace, self-hosted on Postgres.
Turns your task manager into agent memory: hybrid (RRF) retrieval over TickTick or an Obsidian vault via an adapter contract. MCP server + CLI, no vector DB to maintain.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
