get_recent_activityEvent log slice with optional surface filter and keyset pagination on answers.id.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | ||
| cursor | No | ||
| surface | No | e.g. 'claude_mcp', 'app', 'cursor_mcp' |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds behavioral context beyond the readOnlyHint annotation by specifying keyset pagination on answers.id and an optional surface filter. This discloses the pagination mechanism and filtering capability, though it does not mention return format or ordering.
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 a single, front-loaded sentence with no redundant words. It efficiently conveys the core purpose, filter option, and pagination mechanism, earning every word.
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 tool has no output schema, and the description does not explain the response shape or how pagination cursors are obtained, leaving ambiguity about the returned event entries. While the optional surface filter and pagination mechanism are covered, the return value remains underspecified.
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 coverage is only 33% (only surface has a description). The description compensates by explaining that surface is a filter and that cursor uses keyset pagination based on answers.id, giving meaningful context. However, limit semantics are only provided by schema min/max, not the description.
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 retrieves a slice of the event log (recent activity) with optional surface filtering and keyset pagination on answers.id. It distinguishes from sibling tools like get_card and get_user_profile by focusing on activity events, using a specific verb-plus-resource structure.
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 its use for reading recent activity events but does not explicitly state when to use it over siblings or when not to. No alternatives are named, so the agent must infer usage from the tool's purpose and name.
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 cleanly separate single-card lookup, batch lookup, due queue, event log, and SRS adjustments. The main overlap is `get_system_instructions` and `get_user_profile`, which both return CEFR level, languages, due count, and weak words, so an agent could mis-select between them.
Every tool follows a consistent lowercase snake_case verb_noun pattern: capture_*, get_*, mark_*, log_*, check_. The verbs are descriptive and predictable, making the set easy to navigate.
13 tools is well within the ideal scope for a language-learning memory/assistant server. Each tool covers a distinct part of the capture, lookup, review, and spaced-repetition workflow without feeling padded.
The core lifecycle is well covered: grammar and vocabulary capture, batch deck checks, card detail, due queue, recent activity, and SRS boosts/penalties. Minor gaps exist: captured grammar mistakes have no retrieval endpoint, and there is no delete/dismiss path for unwanted cards, but agents can work around these.