Single tool with adequate naming and description, but critical gaps in schema documentation and parameter descriptions. Tool name 'list_indexed_sources' follows verb_noun convention. Description (156 chars) is within baseline range (p10=34, p90=392) and explains purpose clearly. However, the input parameter 'status' lacks a description in the visible schema, only an enum and type are shown. No output schema is documented, forcing LLMs to guess the response structure. This violates the pattern:tool requirement that output schemas be documented. The tool is READ_ONLY (good security posture), but lacks the structured output documentation needed for reliable agent composition.

• Document the output schema explicitly. Example: 'Returns an array of source objects, each with: {id: string, displayName: string, kind: enum(github|notion|...), chunkCount: number, status: enum(ready|running|failed|idle)}'.
• Add description to the 'status' parameter: 'Filter results by indexing status. ready = fully searchable; running = currently indexing; failed = last indexing failed; idle = not yet indexed; any = all statuses.'
• If search_code is a real tool, register it with full schema and description. If not, remove the reference from list_indexed_sources description.
• Add pagination support: accept optional 'limit' (default 20, max 100) and 'offset' (default 0) parameters. Return 'total' count in response so agents know if more results exist.
• Expand tool description to clarify the response structure: 'Returns a list of all indexed sources accessible to the authenticated user, including their current indexing status and content volume (chunk count). Use this to discover available sources before calling search_code.'