eyepup_logLog a change you just shipped. The dossier agent reasons about every visitor profiled AFTER this log row, so it can grade whether the friction pattern recovered. CALL THIS after every UX edit.
| Name | Required | Description | Default |
|---|---|---|---|
| kind | No | ||
| site | No | ||
| paths | No | Comma-separated paths affected. | |
| title | Yes | One-line headline of what shipped. | |
| description | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description discloses that the dossier agent will reason about visitors profiled after the log row, affecting friction-pattern recovery grading. This is a meaningful side effect beyond the mere act of logging, and with no annotations, it carries the transparency burden well. It doesn't cover all potential behaviors but is sufficient for a logging action.
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?
It's exactly two sentences, with the main action front-loaded and the contextual explanation in the second sentence. No wasted words.
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 core purpose, when to invoke, and a key behavioral effect, which is strong given no annotations or output schema. However, it omits guidance on parameter values (e.g., what 'kind' options mean, how to specify sites/paths), so it's not fully complete for a 5-param tool.
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?
The description adds no details about the five input parameters. With only 40% schema coverage (paths and title have descriptions), the remaining parameters (kind, site, description) are undocumented in both schema and description, leaving the agent to infer their meanings. The description does not compensate for this gap.
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 opens with the imperative 'Log a change you just shipped,' clearly identifying the action and object. The title reinforces this. It distinguishes from siblings by focusing on logging after a UX edit, unlike ask/todo/visitor tools.
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?
It explicitly states 'CALL THIS after every UX edit,' providing a concrete trigger condition. It doesn't mention alternatives or when not to use it, but the context of siblings makes this clear enough; the instruction is unambiguous about the intended use case.
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.
Each tool has a clearly distinct purpose: ask is for free-text questions, log is for recording shipped changes, todo is for top friction patterns, visitor is for a specific visitor profile, and visitors_hot is for active high-intent visitors. There is minimal overlap, and the descriptions explicitly guide when to use each.
All tool names share the 'eyepup_' prefix, but the second part mixes verbs (ask, log) with nouns (todo, visitor, visitors_hot). This is readable but not a consistent verb_noun convention. A more uniform pattern like 'ask_question', 'log_change', 'get_todo', 'get_visitor', 'list_hot_visitors' would be clearer.
With 5 tools, the server is well-scoped for its purpose of visitor behavior analysis and optimization. Each tool addresses a distinct need without unnecessary bloat or overlap. The count feels right for a focused analytics assistant.
The toolset covers the core workflow: asking questions, identifying friction patterns, viewing hot visitors, drilling into individual visitor profiles, and logging changes for impact analysis. Minor gaps exist, such as no direct tool for listing all visitors or viewing historical trend charts, but the 'ask' tool can likely address many such queries. Overall, the lifecycle is adequately covered for the stated purpose.