
from mnemonic
Search and synthesize across mnemonic memory files (namespaces, tags, types, full-text) to answer questions and surface sources.
Provides iterative, progressive-disclosure search across a filesystem-backed mnemonic memory store. Use it to run fast regex or semantic searches, filter by namespace/tag/type/date, and synthesize findings into a concise answer with source citations. It supports both simple frontmatter lookups and more advanced semantic queries when qmd tooling is available.
Use this skill when a user asks to "search memories", "find in memories", "grep mnemonic" or requests a deep synthesis of stored notes (e.g., "what do we know about X"). Start with title-only scans, escalate to frontmatter previews, and only read full files when necessary. It also supports iterative refinement for complex queries.
rg, find, and qmd for BM25/vector search if available.Best with agents that can run shell tools and read repo files (Claude Code, Copilot-like assistants, local CLI agents). Supports semantic tooling (qmd) when available for enhanced vector search.
Mnemonic Search provides structured search across mnemonic memory files using ripgrep and optional qmd semantic search. No bundled scripts — purely instruction-based. Progressive disclosure protocol (titles → frontmatter → full detail) is well-designed. Shell injection risk from unquoted ${MNEMONIC_ROOT} variable in bash examples.
Clean skill, no security concerns beyond minor shell variable quoting. Niche audience — only useful for mnemonic ecosystem users.
Mnemonic Setup
Installs and configures a filesystem-backed mnemonic memory system for Claude-style agents, creating directory structure, CLAUDE.md hooks, and an initial contex
Mnemonic — Memory & Capture Core
Agent memory/capture workflow: triggers, silent-capture protocol, minimal memory format, and recall/search commands for saving and retrieving structured agent m