RepoLens provides 7 tools with basic schemas and descriptions, but falls short of production quality. All tools have descriptions (10-80 chars, mostly under the 194-char baseline), and input schemas are present with type definitions. However, descriptions are terse and lack context for LLM selection. The 'edit' tool accepts a hash parameter for verification but lacks error guidance on hash mismatch. Output schemas are not documented, LLMs cannot predict response structure. Parameter descriptions are minimal (e.g., 'Search query string' for 'search' lacks guidance on syntax, limits, or expected result count). No pagination support despite tools like 'search' and 'symbol' returning lists. Error handling is absent, no recovery guidance or actionable error messages. Tool composition is reasonable (each does one thing), but missing chaining IDs in responses (e.g., 'search' should return file paths that 'read' accepts). Overall, this is a functional but under-documented toolkit typical of early-stage projects.
Get dependencies for a file
Edit file contents with hash verification
Get symbol outline for a file
Get reverse dependencies for a file
Read file contents with optional line range and detail level
Search the codebase using trigram-based full-text search
Find symbols by name across the codebase
Output schemas not documented. LLMs cannot predict response structure (fields, types, nesting) for any tool. This forces agents to guess at downstream field names and risks failed tool chains.
Descriptions are terse (10-45 chars) and lack context. 'Search the codebase using trigram-based full-text search' does not explain when to use search vs symbol, what syntax is supported, or result limits. Baseline is 194 chars.
No pagination support. 'search' and 'symbol' accept a 'limit' parameter but no offset/cursor or total_count in response. Large result sets will blow context windows without pagination.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 40 | 2026-07-28+ | v2 |
Error handling absent. 'edit' tool accepts a hash for verification but provides no guidance on hash mismatch errors, file-not-found recovery, or permission failures. Agents cannot self-correct.
Parameter descriptions are minimal. 'lines' in 'read' is described as 'Optional line range (e.g., '10-20')' but lacks format specification, validation rules, or behavior when range exceeds file length.