CPersona
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CPERSONA_RRF_K | No | RRF smoothing parameter | 60 |
| CPERSONA_DB_PATH | No | SQLite database path | ./cpersona.db |
| CPERSONA_SEARCH_MODE | No | Search strategy (`rrf` or `cascade`) | rrf |
| CPERSONA_EMBEDDING_URL | No | Embedding server URL | http://127.0.0.1:8401/embed |
| CPERSONA_AUTO_CALIBRATE | No | Auto-calibrate on startup | false |
| CPERSONA_EMBEDDING_MODE | No | Embedding mode (`http` or `disabled`) | http |
| CPERSONA_CONFIDENCE_ENABLED | No | Include confidence metadata in results | false |
| CPERSONA_TASK_QUEUE_ENABLED | No | Enable background task queue | false |
| CPERSONA_VECTOR_SEARCH_MODE | No | Vector search mode | remote |
Capabilities
Features and capabilities supported by this server
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| storeA | Store a message in agent memory for future recall. |
| recallA | Recall relevant memories using multi-strategy search (vector + FTS5 + keyword). |
| get_profileB | Get the current profile for an agent. |
| update_profileB | Save a pre-computed agent profile to the database. |
| archive_episodeA | Archive a conversation episode with pre-computed summary, keywords, and resolved status. All LLM processing is performed by the caller. |
| list_memoriesB | List recent memories for an agent (for dashboard display). |
| list_episodesA | List archived episodes for an agent (for dashboard display). |
| delete_agent_dataA | Delete ALL data (memories, profiles, episodes) for a specific agent. Used by kernel during agent deletion. |
| calibrate_thresholdA | Auto-calibrate vector search threshold using null distribution z-score. Samples random memory pairs, computes cosine distribution, sets threshold at mean + z*std. No labels used, purely statistical. Adapts to both embedding model and corpus characteristics. |
| delete_memoryA | Delete a single memory by ID. Ownership is enforced when agent_id is provided. |
| delete_episodeA | Delete a single episode by ID. Ownership is enforced when agent_id is provided. |
| get_queue_statusA | Get the status of the background task queue (pending tasks, retry config). |
| export_memoriesA | Export memories, episodes, and profiles to a JSONL file for backup or portability. |
| import_memoriesA | Import memories, episodes, and profiles from a JSONL file. Idempotent via msg_id deduplication. |
| merge_memoriesA | Merge memories, episodes, and profiles from one agent into another. Atomic one-shot equivalent of export→import without intermediate files. Strategy 'skip' deduplicates by msg_id (memories) and summary (episodes). |
| check_healthA | Check memory database health (15 checks). Detects contamination, duplicates, oversized content, embedding issues, FTS desync, invalid JSON/timestamps, stale tasks, missing profiles. Returns storage stats. Set fix=true to auto-repair. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Cloto-dev/CPersona'
If you have feedback or need assistance with the MCP directory API, please join our Discord server