ferdousbhai

Recall For Omarchy — Development skill for Claude Code

Development community

Local multimodal semantic search for Omarchy, with a recall CLI and bundled agent skill.

How to install Recall For Omarchy

This entry records only its repository, not the path inside it, so there is no exact command to give. Open ferdousbhai/recall-for-omarchy and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Recall For Omarchy does

Local multimodal semantic search for Omarchy, with a recall CLI and bundled agent skill.

Alternatives in Development

Mempalace Search — Search your memories across the MemPalace using semantic search with wing/room filtering 58.6k ★
Serena — Code Intelligence Advanced code navigation and semantic search 14k ★
Coding Agent Session Search — Unified TUI and CLI to index and search your local coding agent session history across 11+ providers (Codex, C 708 ★

README

Recall for Omarchy

Search your files by what’s in them. A native Omarchy desktop app **with a CLI and a bundled skill, so coding agents can search your files too**.

Describe a photo, find a document passage, jump to a recording interval, or search source code by intent. EmbeddingGemma 2 runs locally; file content, OCR text and embeddings stay on your computer.

Install

curl -fsSL https://ferdousbhai.com/recall/install.sh | bash

The installer asks before installing packages (`bash -s -- --yes` or `RECALL_YES=1` skips the prompt). It adds a marked Super+Ctrl+Shift+F block to `~/.config/hypr/bindings.lua` after backing it up, and restores the file if Hyprland reports config errors; set `RECALL_NO_KEYBINDING=1` to skip it. If the GitHub API is rate limited, set `GITHUB_TOKEN` or pin `RECALL_VERSION=v0.4.2`.

Release assets from v0.4.3 on are signed with Sigstore (keyless, from this repo's release workflow). To verify a download:

cosign verify-blob recall-for-omarchy-x86_64.tar.gz \
  --bundle recall-for-omarchy-x86_64.tar.gz.sigstore.json \
  --certificate-identity-regexp '^https://github.com/ferdousbhai/recall-for-omarchy/.github/workflows/release.yml@refs/tags/v' \
  --certificate-oidc-issuer https://token.actions.githubusercontent.com

Supports Omarchy on x86_64 and aarch64. The installer downloads a versioned, checksum-verified release, installs runtime dependencies through Omarchy, and configures the desktop integration and agent skill. Vulkan GPU acceleration is included on both architectures, with CPU fallback. Apple Silicon on Linux uses Asahi’s Honeykrisp driver; the installer adds `vulkan-asahi` on Apple machines. Embedding model files (about 865 MB: a 310 MB text model and a 555 MB image/audio encoder) download once from Hugging Face at a pinned revision, after you choose folders to index, and are SHA-256 verified. That download and the install itself are the only network access; search and indexing run offline. Set `RECALL_O