Coding agents

HydraDB plugins give coding agents memory that lasts across sessions: they recall relevant notes when you send a prompt, sync your workspace docs into HydraDB, and save conversations as durable memories. Every plugin on this page runs on the same engine, so they share the commands, modes, and config variables at the end of the page. Each needs an API key and a database ID from the HydraDB dashboard, set as HYDRADB_API_KEY and HYDRADB_DATABASE.

Claude Code

Codex

Cursor

OpenCode

Muse Code

Grok Build

To use HydraDB from any other MCP client, connect the MCP server instead.

Claude Code

Install the plugin

Run these commands inside Claude Code:

/plugin marketplace add hydra-db/hydradb-claude-code
/plugin install hydradb@hydradb
/reload-plugins

Get credentials

Create an API key from the HydraDB dashboard
Create or copy your database ID from the HydraDB dashboard

Set environment variables

macOS/Linux (bash)
macOS/Linux (zsh)
Windows (PowerShell)

echo 'export HYDRADB_API_KEY="your-api-key"' >> ~/.bashrc
echo 'export HYDRADB_DATABASE="your-database"' >> ~/.bashrc
echo 'export HYDRADB_COLLECTION=""' >> ~/.bashrc
source ~/.bashrc

echo 'export HYDRADB_API_KEY="your-api-key"' >> ~/.zshrc
echo 'export HYDRADB_DATABASE="your-database"' >> ~/.zshrc
echo 'export HYDRADB_COLLECTION=""' >> ~/.zshrc
source ~/.zshrc

[System.Environment]::SetEnvironmentVariable("HYDRADB_API_KEY", "your-api-key", "User")
[System.Environment]::SetEnvironmentVariable("HYDRADB_DATABASE", "your-database", "User")
[System.Environment]::SetEnvironmentVariable("HYDRADB_COLLECTION", "", "User")

To verify the values are set:

macOS/Linux
Windows (PowerShell)

echo $HYDRADB_API_KEY
echo $HYDRADB_DATABASE

echo $env:HYDRADB_API_KEY
echo $env:HYDRADB_DATABASE

Run setup and verify

/hydradb:setup
/hydradb:status

HYDRADB_COLLECTION must be set explicitly. To use the default collection, set it to an empty string ("") rather than omitting it. These names need plugin 1.1.0 or later. The older names HYDRA_DB_API_KEY, HYDRADB_TENANT_ID, and HYDRADB_SUB_TENANT_ID still work, but print a deprecation warning.

HYDRADB_COLLECTION strategy:

Use a workspace-specific value (e.g., claude-my-project) for project-isolated memory.
Use a shared value to carry memory across all sessions and workspaces.

Once configured, the plugin runs in the background: it syncs workspace docs on session start, recalls relevant context before each prompt, and captures conversation memory after each response.

Codex

The Codex plugin recalls relevant notes when you send a prompt, syncs your markdown docs into HydraDB, and saves conversations as durable memories, the same behavior and the same engine as the Claude Code plugin above. The JSON config keys, capture/search/ingest modes, and environment variables are identical; only the install flow differs.

Install the plugin

git clone https://github.com/hydra-db/hydradb-codex-plugin.git
cd hydradb-codex-plugin
make bootstrap

Requires Node.js 18+ and npm. Then add a local marketplace entry in your project’s .agents/plugins/marketplace.json pointing at this plugin folder and enable it in .codex/config.toml. See the Codex plugin docs for the current install flow.

Get credentials

Create an API key from the HydraDB dashboard
Create or copy your database ID from the HydraDB dashboard

Set environment variables

export HYDRADB_API_KEY="your-api-key"
export HYDRADB_DATABASE="your-database"
export HYDRADB_COLLECTION=""

Set HYDRADB_COLLECTION to an empty string for the default collection, or a workspace-specific value for project-isolated memory.

The Commands, Modes, and Config Variables below apply to Codex unchanged.

Cursor

Persistent, cross-session memory for Cursor: recalls relevant notes on each prompt, syncs your markdown docs into HydraDB, and saves conversations as durable memories.

Bootstrap & set credentials

git clone https://github.com/hydra-db/hydradb-cursor-plugin.git
cd hydradb-cursor-plugin
make bootstrap
export HYDRADB_API_KEY="your-api-key"
export HYDRADB_DATABASE="your-database"

Requires Node.js 18+ and npm.

Install in Cursor

Install via Cursor’s plugin/marketplace flow pointing at this checkout.

Repo: hydradb-cursor-plugin.

OpenCode

Native plugin for OpenCode that hooks the session lifecycle to recall context at prompt time, sync workspace docs, and capture turns.

OpenCode surface	Engine action
`hydradb_recall` tool (agent-invocable)	recall memory/knowledge for a query
`event`: `file.edited`	incremental workspace sync
`event`: `session.idle`	capture the completed turn

Install & set credentials

Add the plugin to your opencode.json; OpenCode installs it from npm:

{ "plugin": ["hydradb-opencode-plugin"] }

export HYDRADB_API_KEY="your-api-key"
export HYDRADB_DATABASE="your-database"

Requires Bun (OpenCode’s runtime) and Node.js 18+.

Run with tool approval

Tool calls need approval. Run with --auto or approve hydradb_recall interactively:

opencode run --auto "What did we decide? Use hydradb_recall."

Repo: hydradb-opencode-plugin. Package: hydradb-opencode-plugin.

Muse Code

Memory for Muse Code, Meta’s terminal coding agent. Muse Code’s hooks cannot inject context back to the model, so recall is agent-invoked through MCP or skills, while hooks handle automatic capture and doc sync.

Extension point	Role
MCP (`mcp_servers`)	recall (`hydradb_query`) + ingest
Skills (`.claude/skills/`)	`/hydradb …` slash commands (manual recall / remember / sync)
Hooks (`.muse/hooks.json`)	automatic capture (`Stop`) + doc sync (`SessionStart`, `PostToolUse`)

Add the MCP server

{
  "mcp_servers": {
    "hydradb": {
      "transport": "streamable_http",
      "url": "https://mcp.hydradb.com",
      "headers": {
        "Authorization": "Bearer ${HYDRADB_API_KEY}",
        "X-HydraDB-Database": "your-database"
      },
      "enabled": true,
      "mode": "optional"
    }
  }
}

API-key requests to the MCP server require the X-HydraDB-Database header (your database ID). See the MCP reference for all supported headers.

Set credentials & run with the required flags

export HYDRADB_API_KEY="your-api-key"
export HYDRADB_DATABASE="your-database"

# --trust-workspace loads project skills + hooks; --yolo (or an approval
# profile) lets the auto-recall skill run node; both are required or recall
# silently no-ops in headless mode.
muse exec --trust-workspace --yolo "what did we decide about X?"

Only use --yolo in a workspace you trust. It bypasses command approval, so any project hooks loaded by --trust-workspace run against your local environment. In untrusted checkouts, prefer an approval profile that permits only the recall skill’s node call instead of --yolo.

Repo: hydradb-muse-code-plugin.

Grok Build

Automatic long-term memory for Grok Build, xAI’s coding agent. Grok Build’s plugin system is Claude-Code-compatible, so this is a full plugin with real per-prompt auto-recall (not just MCP tools).

Grok Build hook	Engine action
`SessionStart`	print status + async workspace sync
`UserPromptSubmit`	recall memory/knowledge, inject as `additionalContext`
`PostToolUse` (file writes)	incremental workspace sync
`Stop`	capture the completed turn

Set credentials

export HYDRADB_API_KEY="your-api-key"
export HYDRADB_DATABASE="your-database"

Install in Grok Build

Use /marketplace to browse and install (from a local checkout, or via the catalog once listed).

MCP alternative: grok mcp add hydradb --url https://mcp.hydradb.com. The plugin already gives automatic recall, so MCP is optional. Repo: hydradb-grok-plugin.

Everything from here on applies to every plugin on this page.

Commands

Command	Description
`/hydradb:setup`	Interactive configuration wizard
`/hydradb:status`	Show current config and connection status
`/hydradb:search <query>`	Search HydraDB manually
`/hydradb:remember <note>`	Save a note to memory
`/hydradb:save-session`	Manually save the full session to HydraDB
`/hydradb:sync-workspace [--force]`	Force re-sync workspace docs

Modes

`captureMode`: how conversations are saved

Value	Behavior
`session-upsert`	(default) Maintains one evolving session transcript, upserted after each response
`turn`	Saves each user/assistant exchange as an isolated memory
`both`	Saves isolated turns and a rolling session transcript
`off`	No automatic saves; manual saves still work via `/hydradb:save-session`

`searchMode`: what is recalled at prompt time

Value	Behavior
`memory`	(default) Recalls from Memories only
`knowledge`	Recalls from Knowledge only
`both`	Recalls from both in parallel, then merges results

`ingestionMode`: where workspace docs are synced

Value	Behavior
`memory`	(default) Syncs workspace docs into Memories using `infer`
`knowledge`	Syncs workspace docs into the Knowledge base
`auto`	Prefers memory; falls back to knowledge for larger files. Pair with `searchMode: "both"`

`recallMode`: speed vs. depth

Value	Behavior
`fast`	(default) Lower latency, standard recall
`thinking`	Deeper reasoning-based recall via graph traversal

By default, HydraDB syncs each conversation pair (user and assistant) that does not include the ignoreMarker (hydra-ignore). For manual-only capture, set captureMode to off and use /hydradb:save-session or /hydradb:remember <text>.

Config Variables

Required

Variable	Env var	Description
`apiKey`	`HYDRADB_API_KEY`	Your HydraDB API key
`tenantId`	`HYDRADB_DATABASE`	Your HydraDB database
`subTenantId`	`HYDRADB_COLLECTION`	Collection scope. Set to `""` for the default collection

Core behavior

Variable	Default	Description
`autoRecall`	`true`	Inject HydraDB context before every prompt automatically
`autoIngest`	`true`	Sync workspace docs on session start and file edits automatically
`captureMode`	`session-upsert`	How conversations are saved (see Modes)
`searchMode`	`memory`	What to recall at prompt time (see Modes)
`ingestionMode`	`memory`	Where workspace docs are synced (see Modes)
`recallMode`	`fast`	Recall depth: `fast` or `thinking`
`graphContext`	`true`	Include entity paths and graph relations in recalled context

Limits

Variable	Default	Description
`maxContextChars`	`7000`	Max characters injected into the prompt from HydraDB
`maxMemoryResults`	`6`	Max memory chunks returned per recall
`maxKnowledgeResults`	`4`	Max knowledge chunks returned per recall
`maxFileSizeBytes`	`52428800` (50 MB)	Max file size for workspace sync
`maxFilesPerSync`	`25`	Max files synced per pass
`maxMemoryCharsPerChunk`	`52428800` (50 MB)	Max chars per memory chunk
`maxMemoryChunksPerFile`	`1`	Max chunks per file

Timeouts

Variable	Env var	Default	Description
`requestTimeoutMs`	`HYDRADB_REQUEST_TIMEOUT_MS`	`15000`	Timeout for recall/read requests (ms)
`writeTimeoutMs`	`HYDRADB_WRITE_TIMEOUT_MS`	`15000`	Timeout for sync/write requests (ms)

File filtering

Variable	Default	Description
`includeGlobs`	`CLAUDE.md`, `.claude/*/.md`, `*/.md`, `*/.mdx`	Files included in workspace sync
`excludeGlobs`	`.git/`, `node_modules/`, `dist/`, `/.env`, `/.key`, etc.	Files excluded from workspace sync
`ignoreMarker`	`hydra-ignore`	Add to a file or prompt to skip capture and sync

By default, HydraDB includes .md and .mdx files in workspace sync. To include additional file types, extend includeGlobs in your config file.

Advanced

Variable	Default	Description
`subTenantId`	-	Collection: a workspace-level namespace within your database
`userName`	-	Optional user name attached to memories
`apiBaseUrl`	`https://api.hydradb.com`	HydraDB API base URL
`memoryCustomInstructions`	-	Custom instructions for how conversation memories are extracted
`workspaceMemoryCustomInstructions`	-	Custom instructions for how workspace doc memories are extracted
`debug`	`false`	Enable debug logging to `.hydradb-plugin-data/debug.log`

Config File Location Priority

The plugin resolves config in this order (later sources override earlier ones):

Built-in defaults
HYDRADB_PLUGIN_CONFIG env var (absolute path to a JSON file)
${CLAUDE_PLUGIN_DATA}/config.json
.hydradb-plugin-data/config.json
.hydradb-plugin.json (workspace-shared, safe to commit to git)
.hydradb-plugin.local.json (workspace-local, add to .gitignore)
Environment variable overrides

Suggested Modes

Use case	`captureMode`	`searchMode`	`ingestionMode`
Default / everyday use	`session-upsert`	`memory`	`memory`
Cross-session continuity	`session-upsert`	`memory`	`memory`
Team shared knowledge base	`session-upsert`	`memory`	`memory`
Maximum recall coverage	`both`	`both`	`auto`
Recall only, no auto-save	`off`	`memory`	`memory`
Isolated turn snapshots	`turn`	`memory`	`memory`

Suggested Configs

Minimal (env vars only)

export HYDRADB_API_KEY="your-api-key"
export HYDRADB_DATABASE="your-database"
export HYDRADB_COLLECTION=""

Recommended workspace config (`.hydradb-plugin.json`)

{
  "apiKey": "${HYDRADB_API_KEY}",
  "tenantId": "${HYDRADB_DATABASE}",
  "subTenantId": "claude-my-workspace",
  "autoRecall": true,
  "autoIngest": true,
  "captureMode": "session-upsert",
  "searchMode": "memory",
  "ingestionMode": "memory",
  "recallMode": "fast",
  "graphContext": true,
  "maxContextChars": 7000,
  "requestTimeoutMs": 15000,
  "writeTimeoutMs": 15000
}

Maximum recall

{
  "captureMode": "both",
  "searchMode": "both",
  "ingestionMode": "auto",
  "maxContextChars": 10000,
  "maxMemoryResults": 8,
  "maxKnowledgeResults": 6
}

Manual-save only (no auto-capture)

{
  "captureMode": "off",
  "autoRecall": true,
  "autoIngest": true
}

Get Started

Usage

Organize

Understand

Integrations

For Agents

Claude Code

Codex

Cursor

OpenCode

Muse Code

Grok Build

Claude Code

Codex

Cursor

OpenCode

Muse Code

Grok Build

Commands

Modes

`captureMode`: how conversations are saved

`searchMode`: what is recalled at prompt time

`ingestionMode`: where workspace docs are synced

`recallMode`: speed vs. depth

Config Variables

Required

Core behavior

Limits

Timeouts

File filtering

Advanced

Config File Location Priority

Suggested Modes

Suggested Configs

Minimal (env vars only)

Recommended workspace config (`.hydradb-plugin.json`)

Maximum recall

Manual-save only (no auto-capture)

Source code

Claude Code

Codex

Cursor

OpenCode

Muse Code

Grok Build

​Claude Code

​Codex

​Cursor

​OpenCode

​Muse Code

​Grok Build

​Commands

​Modes

​captureMode: how conversations are saved

​searchMode: what is recalled at prompt time

​ingestionMode: where workspace docs are synced

​recallMode: speed vs. depth

​Config Variables

​Required

​Core behavior

​Limits

​Timeouts

​File filtering

​Advanced

​Config File Location Priority

​Suggested Modes

​Suggested Configs

​Minimal (env vars only)

​Recommended workspace config (.hydradb-plugin.json)

​Maximum recall

​Manual-save only (no auto-capture)

​Source code

Claude Code

Codex

Cursor

OpenCode

Muse Code

Grok Build

Commands

Modes

`captureMode`: how conversations are saved

`searchMode`: what is recalled at prompt time

`ingestionMode`: where workspace docs are synced

`recallMode`: speed vs. depth

Config Variables

Required

Core behavior

Limits

Timeouts

File filtering

Advanced

Config File Location Priority

Suggested Modes

Suggested Configs

Minimal (env vars only)

Recommended workspace config (`.hydradb-plugin.json`)

Maximum recall

Manual-save only (no auto-capture)

Source code