Wire Memory exposes 2 tools with basic schemas and descriptions, but falls short of production quality. Both tools have descriptions (wire_search: 73 chars, wire_write: 79 chars) and input schemas with typed parameters. However, descriptions lack actionable context (no WHEN to use, no prerequisites, no error recovery guidance). Parameters lack constraints (no enums, ranges, or format specs). Output schemas are completely undocumented, LLMs cannot predict what wire_search returns or how to chain results into wire_write. No error handling guidance. Tool names are verb-noun (good), but descriptions are generic and do not explain dependencies or side effects. Missing: pagination for search results, idempotency guarantees for write, and field-level documentation.
Search wire-memory for previous decisions, patterns, or discussions
Log decisions, patterns, corrections, and conventions to wire-memory
Output schemas completely undocumented. wire_search returns unknown structure; wire_write returns unknown confirmation. LLMs cannot predict response fields or chain results.
Descriptions lack actionable context. No WHEN to use, no prerequisites, no error recovery guidance. 'Search wire-memory for previous decisions' does not explain what happens if no results match, or how to interpret results.
Parameters lack constraints. 'query' and 'content' are free-form strings with no length limits, format specs, or validation rules. No enums for 'project' scope.
No pagination or result limits documented. wire_search may return unbounded results, risking context window exhaustion. No limit, offset, or cursor parameters.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 52 | 2026-07-28+ | v2 |
wire_write lacks idempotency guarantees and confirmation step. No dry-run mode. Agents cannot safely retry on ambiguous failures without risking duplicate entries.