Call first. Returns the active project, repositories, and index health in one call, plus a usage field (decision matrix + routing). Non-empty repositories = proceed with graph queries; empty = tell the user to attach repos on the dashboard and stop, do not fall back to Grep.
| Name | Required | Description | Default |
|---|---|---|---|
| feedback | No | Optional: report how a previous answer worked out; piggybacks on any call. |
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?
Annotations already declare readOnly, idempotent, and non-destructive behavior. The description adds useful behavioral context beyond those annotations, such as returning a usage field with decision matrix/routing and the explicit stop condition for empty repositories. It does not contradict the annotations.
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?
Three tight sentences with no filler. The key instruction 'Call first' is front-loaded, followed by the return summary and the conditional routing rule. Every sentence earns its place.
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?
For a zero-required-parameter initialization tool, the description is complete: it states what is returned, how to interpret the result, what to do in the empty case, and explicitly warns against falling back to Grep. The lack of an output schema is compensated by the clear high-level return contract.
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 sole parameter, feedback, is fully documented in the schema with descriptions for answerId, outcome, and correction, so the schema carries the parameter burden. The description adds no parameter-specific meaning, but it does not need to because coverage is 100% and the parameter is optional.
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 states a clear purpose: initialize the session by returning active project, repositories, and index health in one call, plus a usage field for routing. This distinguishes it from graph-query siblings by framing it as the mandatory first call and by naming the exact aggregated data it returns.
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 gives explicit usage direction: 'Call first' establishes when to invoke it, and the conditional rule ('Non-empty repositories = proceed with graph queries; empty = tell the user to attach repos... do not fall back to Grep') tells the agent exactly when to continue and when to stop. This is strong routing guidance relative to alternatives.
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.
Tools generally target distinct operations—repository attachment, symbol resolution, graph traversal, source access, indexing health—and the detailed routing notes reduce ambiguity. A few pairs still overlap in intent (quick_lookup vs locate symbol, context vs ask_codebase), so it is not perfectly unambiguous.
All names are lowercase snake_case and readable, but the set mixes verb_noun patterns like add_repository and list_projects with single-word nouns/verbs like context, history, map, relate, and source, plus one acronym (adr). There is no predictable convention across the full surface, though individual names are clear.
25 tools is at the high end for a single MCP server and falls in the 'feels heavy' range. The breadth is justified by the code-intelligence/query domain, but the selection surface is large for an agent to navigate efficiently.
The code indexing, search, graph query, source access, history, ADR, test lookup, and scope estimation needs are well covered. Minor gaps remain—no detach/remove repository, no ADR content editing, and source access is gated by an add-on—but these are workable.