io.github.soluai-spa/solucortex-mcp is an MCP server published by io.github.soluai-spa, hosted at mcp.solucortex.ai. When last checked on 2026-11-22 it was working: it answers and lists its tools. It declares 6 tools, including solucortex_recall, solucortex_search, solucortex_remember, solucortex_update_memory, solucortex_flag_memory.
Living technical memory for AI agents: recall approved project context, record what was learned.
Nothing else in the archive shares these values.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
Inferred from the host name and headers: what a server runs on, not who runs it.
Build living context for a task (POST /context/build). Call this at the START of a task, before touching code: returns approved, active memories (decisions, conventions, risks, sensitive modules, architecture) ranked by semantic similarity + importance. Uses OpenAI embeddings (slower, 20 req/min).
| Parameter | Type | Description |
|---|---|---|
| query * | string | Describe the current task/module in natural language, e.g. 'implement API key rotation in the secrets module'. Used to semantically retrieve the most relevant memories. |
| project_id | any | Project UUID. If omitted, the session default applies, else the backend infers it from the API key. |
{
"name": "solucortex_recall",
"description": "Build living context for a task (POST /context/build).\n\n Call this at the START of a task, before touching code: returns approved, active\n memories (decisions, conventions, risks, sensitive modules, architecture) ranked by\n semantic similarity + importance. Uses OpenAI embeddings (slower, 20 req/min).",
"inputSchema": {
"type": "object",
"title": "solucortex_recallArguments",
"required": [
"query"
],
"properties": {
"query": {
"type": "string",
"title": "Query",
"description": "Describe the current task/module in natural language, e.g. 'implement API key rotation in the secrets module'. Used to semantically retrieve the most relevant memories."
},
"project_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Project Id",
"default": null,
"description": "Project UUID. If omitted, the session default applies, else the backend infers it from the API key."
}
}
},
"outputSchema": {
"type": "object",
"title": "solucortex_recallDictOutput",
"additionalProperties": true
},
"annotations": {
"title": "SoluCortex: recall task context",
"readOnlyHint": true,
"openWorldHint": true
}
}Ad-hoc semantic search of memories (POST /search/semantic). Use for specific questions during a task (e.g. 'how is authentication implemented?'), distinct from recall which builds the full startup context. Uses OpenAI embeddings (20 req/min).
| Parameter | Type | Description |
|---|---|---|
| limit | integer | Max memories to return. |
| query * | string | Question or topic to search across the project's memories. |
| project_id | any | Project UUID. If omitted, the session default applies, else the backend infers it from the API key. |
{
"name": "solucortex_search",
"description": "Ad-hoc semantic search of memories (POST /search/semantic).\n\n Use for specific questions during a task (e.g. 'how is authentication implemented?'),\n distinct from recall which builds the full startup context. Uses OpenAI embeddings\n (20 req/min).",
"inputSchema": {
"type": "object",
"title": "solucortex_searchArguments",
"required": [
"query"
],
"properties": {
"limit": {
"type": "integer",
"title": "Limit",
"default": 10,
"maximum": 50,
"minimum": 1,
"description": "Max memories to return."
},
"query": {
"type": "string",
"title": "Query",
"description": "Question or topic to search across the project's memories."
},
"project_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Project Id",
"default": null,
"description": "Project UUID. If omitted, the session default applies, else the backend infers it from the API key."
}
}
},
"outputSchema": {
"type": "object",
"title": "solucortex_searchDictOutput",
"additionalProperties": true
},
"annotations": {
"title": "SoluCortex: semantic memory search",
"readOnlyHint": true,
"openWorldHint": true
}
}Record a memory in SoluCortex (POST /memories). Call when closing a task or making a relevant technical decision. As an authorized agent (Bearer api_key), the memory is stored with status 'approved' and traced. Never store real secrets: if you find one, record location/type/severity and action taken, with a redacted reference.
| Parameter | Type | Description |
|---|---|---|
| type * | string | Memory type. The backend validates the value. Canonical vocabulary: architecture, decision, risk, convention, bug_history, tech_debt, sensitive_module, learning, external_integration. (Older backends may use technical_decision, historical_bug, current_state, task_closure.) On HTTP 422, retry with th… |
| title * | string | Short, clear title (~80 chars max). |
| content * | string | Full content: what to remember, why it matters, where it applies and what it prevents. NEVER include real secrets (tokens, passwords, .env, credentials). |
| importance | integer | Priority 1-10. Default 5. 6-7 high; 8-9 important decision/risk/convention; 10 critical. |
| project_id | any | Project UUID. If omitted, the session default applies, else the backend infers it from the API key. |
{
"name": "solucortex_remember",
"description": "Record a memory in SoluCortex (POST /memories).\n\n Call when closing a task or making a relevant technical decision. As an authorized\n agent (Bearer api_key), the memory is stored with status 'approved' and traced. Never\n store real secrets: if you find one, record location/type/severity and action taken,\n with a redacted reference.",
"inputSchema": {
"type": "object",
"title": "solucortex_rememberArguments",
"required": [
"type",
"title",
"content"
],
"properties": {
"type": {
"type": "string",
"title": "Type",
"description": "Memory type. The backend validates the value. Canonical vocabulary: architecture, decision, risk, convention, bug_history, tech_debt, sensitive_module, learning, external_integration. (Older backends may use technical_decision, historical_bug, current_state, task_closure.) On HTTP 422, retry with the alternate vocabulary."
},
"title": {
"type": "string",
"title": "Title",
"maxLength": 120,
"description": "Short, clear title (~80 chars max)."
},
"content": {
"type": "string",
"title": "Content",
"description": "Full content: what to remember, why it matters, where it applies and what it prevents. NEVER include real secrets (tokens, passwords, .env, credentials)."
},
"importance": {
"type": "integer",
"title": "Importance",
"default": 5,
"maximum": 10,
"minimum": 1,
"description": "Priority 1-10. Default 5. 6-7 high; 8-9 important decision/risk/convention; 10 critical."
},
"project_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Project Id",
"default": null,
"description": "Project UUID. If omitted, the session default applies, else the backend infers it from the API key."
}
}
},
"outputSchema": {
"type": "object",
"title": "solucortex_rememberDictOutput",
"additionalProperties": true
},
"annotations": {
"title": "SoluCortex: record a memory",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}Correct an existing memory (PATCH /memories/{id}). Agents propose, humans govern: the edit is applied but the memory returns to PENDING until a human re-approves it. ALWAYS tell the user the change awaits their approval in the SoluCortex panel. Status changes are not possible here.
| Parameter | Type | Description |
|---|---|---|
| type | any | Corrected memory type (one of the 9 canonical types). |
| title | any | Corrected title. |
| content | any | Corrected full content. |
| memory_id * | string | UUID of the memory to correct. |
| importance | any | Corrected priority 1-10. |
{
"name": "solucortex_update_memory",
"description": "Correct an existing memory (PATCH /memories/{id}).\n\n Agents propose, humans govern: the edit is applied but the memory returns to\n PENDING until a human re-approves it. ALWAYS tell the user the change awaits\n their approval in the SoluCortex panel. Status changes are not possible here.",
"inputSchema": {
"type": "object",
"title": "solucortex_update_memoryArguments",
"required": [
"memory_id"
],
"properties": {
"type": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Type",
"default": null,
"description": "Corrected memory type (one of the 9 canonical types)."
},
"title": {
"anyOf": [
{
"type": "string",
"maxLength": 120
},
{
"type": "null"
}
],
"title": "Title",
"default": null,
"description": "Corrected title."
},
"content": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Content",
"default": null,
"description": "Corrected full content."
},
"memory_id": {
"type": "string",
"title": "Memory Id",
"description": "UUID of the memory to correct."
},
"importance": {
"anyOf": [
{
"type": "integer",
"maximum": 10,
"minimum": 1
},
{
"type": "null"
}
],
"title": "Importance",
"default": null,
"description": "Corrected priority 1-10."
}
}
},
"outputSchema": {
"type": "object",
"title": "solucortex_update_memoryDictOutput",
"additionalProperties": true
},
"annotations": {
"title": "SoluCortex: update a memory (requires human re-approval)",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}Flag a memory as outdated/incorrect (POST /memories/{id}/flag). The memory is marked for human review in the SoluCortex panel; nothing is deleted or unpublished. ALWAYS tell the user a memory awaits their decision in the panel. Deletion is human-only by design.
| Parameter | Type | Description |
|---|---|---|
| reason * | string | Why it needs review: outdated, incorrect, duplicated... (10-500 chars). |
| memory_id * | string | UUID of the memory to flag. |
| suggested_action | string | What you suggest the human does: update, archive, delete or review. |
{
"name": "solucortex_flag_memory",
"description": "Flag a memory as outdated/incorrect (POST /memories/{id}/flag).\n\n The memory is marked for human review in the SoluCortex panel; nothing is\n deleted or unpublished. ALWAYS tell the user a memory awaits their decision\n in the panel. Deletion is human-only by design.",
"inputSchema": {
"type": "object",
"title": "solucortex_flag_memoryArguments",
"required": [
"memory_id",
"reason"
],
"properties": {
"reason": {
"type": "string",
"title": "Reason",
"maxLength": 500,
"minLength": 10,
"description": "Why it needs review: outdated, incorrect, duplicated... (10-500 chars)."
},
"memory_id": {
"type": "string",
"title": "Memory Id",
"description": "UUID of the memory to flag."
},
"suggested_action": {
"type": "string",
"title": "Suggested Action",
"default": "review",
"description": "What you suggest the human does: update, archive, delete or review."
}
}
},
"outputSchema": {
"type": "object",
"title": "solucortex_flag_memoryDictOutput",
"additionalProperties": true
},
"annotations": {
"title": "SoluCortex: flag a memory for human review",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}List the project's memories without semantic search (GET /memories). Useful for quick inspection/audit or to confirm a just-created memory was stored. Does not consume OpenAI quota.
| Parameter | Type | Description |
|---|---|---|
| limit | integer | Max memories to return. |
| project_id | any | Project UUID. If omitted, the session default applies, else the backend infers it from the API key. |
{
"name": "solucortex_list_memories",
"description": "List the project's memories without semantic search (GET /memories).\n\n Useful for quick inspection/audit or to confirm a just-created memory was stored.\n Does not consume OpenAI quota.",
"inputSchema": {
"type": "object",
"title": "solucortex_list_memoriesArguments",
"properties": {
"limit": {
"type": "integer",
"title": "Limit",
"default": 20,
"maximum": 100,
"minimum": 1,
"description": "Max memories to return."
},
"project_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Project Id",
"default": null,
"description": "Project UUID. If omitted, the session default applies, else the backend infers it from the API key."
}
}
},
"outputSchema": {
"type": "object",
"title": "solucortex_list_memoriesDictOutput",
"additionalProperties": true
},
"annotations": {
"title": "SoluCortex: list memories",
"readOnlyHint": true,
"openWorldHint": true
}
}SoluCortex is the project's living technical memory: approved decisions, conventions, risks, architecture and learnings, governed per project. Follow this workflow: 1. START of a task: call `solucortex_recall` with a description of the task/module BEFORE touching code. Treat the returned memories as current project truth — they outrank older docs and your assumptions. 2. DURING the task: call `solucortex_search` for specific questions (e.g. "how is authentication implemented?") instead of re-deriving answers. 3. CLOSE of a task: evaluate whether something reusable was learned. If so, call `solucortex_remember` with the right `type`: - architecture (how the system is built), decision (choice + why), convention (rule to follow), risk (what can go wrong), bug_history (bug + root cause + fix), tech_debt (known shortcut), sensitive_module (handle with care + why), learning (reusable insight), external_integration (third-party service behavior/gotchas). Set `importance` 1-10 (8+ only for decisions/risks that shape future work). Depending on the project's governance, a memory may be stored as pending until a human approves it — that is expected, do not retry. Corrections and lifecycle (agents propose, humans govern): - To fix or improve an existing memory, call `solucortex_update_memory`. The edit is applied but the memory returns to PENDING: always tell the user they must review and approve the change in the SoluCortex panel (https://solucortex.ai). - If a memory looks outdated, wrong or duplicated, call `solucortex_flag_memory` with a clear reason (optionally suggesting archive/delete). A human decides in the panel. - Agents can never approve, reject or delete memories — deletion is human-only by design. Rules: never store secrets (tokens, passwords, .env values) in a memory — record location/type/severity and the action taken instead. Do not log or echo API keys. One memory per fact; for small corrections prefer `solucortex_update_memory` over creating a near-duplicate.
Names, descriptions and schemas were written by the server and are shown as data.
Pattern matches on what the server declares. They show what it says it can do, not what it does.
None yet. Changes appear after a server's second daily check.