save_brain_stateUPSERT the brain's current conversation messages + last-{reflection,vision,act}-at timestamps so a container restart can pick up where it left off. Messages array is trimmed to the most-recent 256 entries server-side.
| Name | Required | Description | Default |
|---|---|---|---|
| last_act_at_ms | No | ||
| last_vision_at_ms | No | ||
| conversation_messages | Yes | ||
| last_reflection_at_ms | No |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Input schema / additionalPropertiesAdded value: +falseInput schema / properties / last_act_at_ms / anyOfPrevious value: -[
- {
- "maximum": 9007199254740991,
- "minimum": -9007199254740991,
- "type": "integer"
- },
- {
- "type": "null"
- }
-]New value: +[
+ {
+ "type": "integer"
+ },
+ {
+ "type": "null"
+ }
+]Input schema / properties / last_reflection_at_ms / anyOfPrevious value: -[
- {
- "maximum": 9007199254740991,
- "minimum": -9007199254740991,
- "type": "integer"
- },
- {
- "type": "null"
- }
-]New value: +[
+ {
+ "type": "integer"
+ },
+ {
+ "type": "null"
+ }
+]Input schema / properties / last_vision_at_ms / anyOfPrevious value: -[
- {
- "maximum": 9007199254740991,
- "minimum": -9007199254740991,
- "type": "integer"
- },
- {
- "type": "null"
- }
-]New value: +[
+ {
+ "type": "integer"
+ },
+ {
+ "type": "null"
+ }
+]Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds behavioral details beyond annotations: it specifies that the operation is an UPSERT (idempotent write) and that messages are trimmed to 256 entries server-side. Annotations already indicate idempotentHint=true and readOnlyHint=false, so the description complements them without contradiction.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences with zero wasted words. It front-loads the action (UPSERT) and immediately conveys purpose, making it efficient and easy to parse.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The description covers the tool's purpose, key data, and server-side trimming behavior. However, it omits mention of return values or error conditions, and the schema offers no property descriptions. For a simple upsert tool, this is largely adequate but not fully complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%. The description only provides high-level mapping of parameters (conversation messages and three timestamps), but lacks detailed format or constraints for each parameter. It adds basic context but does not fully compensate for the missing schema descriptions.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: to UPSERT conversation messages and timestamps for checkpointing across container restarts. It uses a strong verb and specifies the exact data being saved, effectively distinguishing it from the sibling tool load_brain_state.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for saving state before a container restart, but does not explicitly contrast with alternatives like load_brain_state or provide when-not-to-use guidance. It relies on context rather than direct instruction.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Most tools have clearly distinct purposes, with detailed descriptions differentiating them. However, the high count (60) introduces some overlap among memory-related tools (e.g., recall_nearby_memories, search_memories, recent_memory) and environment inspection tools (look_around, look_at, survey_site), causing minor ambiguity.
The majority of tools follow a consistent verb_noun pattern (e.g., enter_space, create_commitment, recall_nearby_memories). A few names break pattern, like cognitive_boot (adjective_noun) or who_is_here (phrase), but overall the naming is predictable and readable.
With 60 tools, the server exceeds the high end of the typical well-scoped range (3-15). While the domain is complex, many tools could be consolidated (e.g., multiple memory retrieval and building tools), making the surface feel bloated rather than lean and focused.
The tool set covers the core functionalities of the 3D world—spatial navigation, building, memory, commitments, skills, and social interaction—with few obvious gaps (e.g., no direct region deletion or agent interaction beyond chat). The breadth is appropriate for the domain, though some redundancy suggests minor over-engineering.