hamyad 0.2.1: one-line install + setup wizard

One-line install + setup wizard

• curl -fsSL https://mrzroot.github.io/hamyad/install.sh | sh (macOS/Linux) and irm https://mrzroot.github.io/hamyad/install.ps1 | iex (Windows): uses Node.js 20+ if present, otherwise a checksum-verified portable Node 24 LTS under ~/.hamyad; installs the latest release, puts hamyad on PATH, asks which project to set up and runs the wizard. Works from a pipe (questions go to the terminal); HAMYAD_YES=1 for unattended installs.
• hamyad setup (alias hamyad tunnel): init if needed, pre-approve hamyad in Claude Code (settings.local.json + project trust), Codex (trusted project), Gemini CLI (trusted folder) and the Cursor CLI, then choose how chat apps reach the brain: free Cloudflare quick tunnel (no account; cloudflared fetched if missing), --worker (deploys your own Cloudflare Worker via wrangler and sets its secrets), --url (your own), or --no-chat. Prints connector URLs with direct links for claude.ai (pre-filled), ChatGPT, Grok, Perplexity and Gemini; --open opens them. The access token lives in .git/hamyad/remote.json, never in the repo.
• Every release also ships hamyad.tgz, so releases/latest/download/hamyad.tgz always points at the newest build.

Site

• One-click copy install button (auto-detects Windows vs macOS/Linux) and an animated EN/فارسی step-by-step tutorial for Claude.ai, ChatGPT and Cursor.

CI

• Installer end-to-end on Ubuntu and macOS (clean PATH without Node, pseudo-terminal answers, re-run, home-directory guard), Windows PowerShell 5.1 and 7, a real Cloudflare quick tunnel smoke test, and setup --worker against a real wrangler deploy --dry-run.

v0.2.0 — one shared brain for every AI

0.2.0 (2026-10-09)

hamyad is now one shared brain for every AI tool, not just Claude.

Connect any AI

• hamyad init --all (or --tools …) wires MCP + hooks + instructions for Claude Code, Codex, Gemini CLI, Cursor, Copilot (VS Code, CLI, cloud agent), Windsurf/Devin Desktop, Zed, Roo Code, JetBrains Junie and aider; --global adds Claude Desktop, Codex user config and Windsurf's global MCP.
• hamyad connect <tool>: exact setup for ChatGPT (MCP connector and Custom GPT Action), claude.ai, Claude Desktop, Grok, Perplexity, Gemini app, Cline, JetBrains AI Assistant and "anything else".
• REST API + OpenAPI 3.1 (/api/*, /openapi.json) on hamyad serve and the Worker, for GPT Actions and tools without MCP; GET /api/memory.md for tools that read a URL.
• ChatGPT gets OpenAI's read-only search/fetch tools (deep research compatible). Writes are attributed per client (?source=, MCP clientInfo, User-Agent).

Capture across tools

• One hook entry point, hamyad hook <tool> <event>, for Claude Code, Codex, Gemini CLI, Cursor, Copilot and Windsurf (payloads verified against the real CLIs). Fail-open; errors go to a local log.
• Every agent session becomes a sessions/ entry and its git diff a changes/ entry (change journal; working-tree snapshots, generated files excluded, dibs attribution when present).
• hamyad import: transcript importers for Claude Code, Codex, Gemini CLI, Copilot CLI, Cursor agent transcripts, Windsurf and aider.
• hamyad timeline: chats, change sets, decisions, supersessions, tasks, agent-attributed git commits and dibs edits.

Decisions that change

• supersedes on brain_remember / --supersedes / hamyad supersede; superseded entries are linked both ways and listed under "Superseded — do NOT follow".
• Conflict hints when a new decision overlaps one still in force.
• Per-tool "last seen": every agent gets "⚠ DECISIONS CHANGED since your last session" at session start, and a mid-session notice on the next prompt.

No-GitHub mode

• hamyad init --store PATH keeps the brain in any synced folder or a single MEMORY.md file (Dropbox, Google Drive, iCloud, a cowork folder); .hamyad.json points the repo at it.
• .brain/MEMORY.md export regenerated on every write (CLI, MCP, hooks, Worker), re-importable; hamyad export.

Triggers and privacy

• Documented triggers: session start, prompt, turn end, session end, MCP/REST write, git post-commit, hamyad watch timer, manual sync.
• Secret redaction on everything written; capture.sessions off | summary (default) | full (opt-in redacted transcripts); size limits; per-tool switches; excludes.

Tests: 54 unit/integration tests plus scripts/e2e (46 checks) running the real Claude Code, Codex, Gemini CLI, Copilot CLI and aider binaries against a mock model server.

Breaking: hook commands are now hamyad hook claude SessionStart etc. (hamyad hook session-start|session-end still work; re-run hamyad init to update). Session records moved from .brain/.state.json to .git/hamyad/.

v0.1.0

Changelog

0.1.0 (2026-10-09)

First release.

• .brain/ git-tracked memory: decisions, tasks, notes, context, sessions as plain Markdown with frontmatter.
• Dependency-free MCP server core with 7 tools (brain_context, brain_remember, brain_search, brain_list, brain_get, brain_update, brain_log_session), a brain://brief resource and two prompts.
• Transports: stdio (Claude Code, Claude Desktop) and stateless Streamable HTTP (claude.ai custom connector), the latter on Node (hamyad serve) and Cloudflare Workers (worker/).
• GitHub backend: one GraphQL read per request, one commit per write, sha-guarded updates.
• Claude Code hooks: SessionStart (fast-forward pull, regenerate CLAUDE.md block, inject what changed on the chat side) and SessionEnd (LLM-free session summary from the transcript, commit, safe push).
• CLI: init, status, sync, add, list, search, show, done, update, context, absorb, connect, mcp, serve, hook.
• Persian-aware search (Arabic/Persian letters, ZWNJ, digits).
• 36 tests, including interop with the official MCP TypeScript SDK client over stdio and Streamable HTTP.