@intent-systems/nexus 2026.1.5-3

package/skills/agent-browser/docs/first-tests.md

@@ -0,0 +1,261 @@
+# Agent-browser demo playbook
+
+This playbook gives two demos: a fast visible check and a harder "solver" flow.
+Both avoid API keys and OAuth.
+
+## Immediate value
+
+- Real browser automation on dynamic sites (not just HTML fetches).
+- Deterministic element refs from snapshots (stable, less brittle selectors).
+- Quick smoke tests for links, forms, and navigation.
+
+## Speed / model choice
+
+- Default to Claude Haiku 4.5 (use the latest Haiku in Claude Code).
+- Claude Code is the preferred dispatcher for fast runs.
+- Wordle uses a local word list; no answer lookup.
+
+## Prereqs
+
+```bash
+agent-browser --version
+agent-browser install
+```
+
+## Test 1: Hacker News headline (easy, visible)
+
+Goal: show a real page open, click, and read content.
+
+```bash
+# Visible browser window
+agent-browser --session demo open https://news.ycombinator.com --headed
+
+# Get interactive elements (refs)
+agent-browser --session demo snapshot -i -c -d 3
+
+# Pick the first story link ref from snapshot (example: @e1)
+agent-browser --session demo click @e1
+
+# Read the new page title and URL
+agent-browser --session demo get title
+agent-browser --session demo get url
+
+# Optional screenshot
+agent-browser --session demo screenshot /tmp/agent-browser-hn.png
+
+agent-browser --session demo close
+```
+
+Success looks like:
+- A headed browser opens.
+- Snapshot lists interactive links with refs.
+- Title/URL change after the click.
+
+## Test 2: NYT Wordle solver (harder, real site)
+
+Goal: solve Wordle using only a local word list and math, plus browser control.
+
+### Approach
+
+1. Keep a local candidate list of valid words.
+2. After each guess, read tile evaluations (green/yellow/gray).
+3. Filter candidates by the feedback.
+4. Pick the next guess by maximizing letter information.
+
+### One-time setup: word list + solver
+
+```bash
+curl -fsSL https://raw.githubusercontent.com/tabatkins/wordle-list/main/words \
+  -o /tmp/wordle-words.txt
+```
+
+Create `~/nexus/home/projects/wordle-solver.mjs`:
+
+```js
+import fs from "fs";
+
+const WORDS_PATH = process.env.WORDLE_WORDS ?? "/tmp/wordle-words.txt";
+const STATE_PATH = process.env.WORDLE_STATE ?? "/tmp/wordle-state.json";
+
+if (!fs.existsSync(WORDS_PATH)) {
+  console.error(
+    "Missing word list. Download with: curl -fsSL https://raw.githubusercontent.com/tabatkins/wordle-list/main/words -o /tmp/wordle-words.txt",
+  );
+  process.exit(1);
+}
+
+const words = fs.readFileSync(WORDS_PATH, "utf8").trim().split(/\s+/);
+
+function score(guess, answer) {
+  const result = Array(5).fill("b");
+  const counts = {};
+  for (let i = 0; i < 5; i++) {
+    const g = guess[i];
+    const a = answer[i];
+    if (g === a) {
+      result[i] = "g";
+    } else {
+      counts[a] = (counts[a] ?? 0) + 1;
+    }
+  }
+  for (let i = 0; i < 5; i++) {
+    if (result[i] === "g") continue;
+    const g = guess[i];
+    if ((counts[g] ?? 0) > 0) {
+      result[i] = "y";
+      counts[g]--;
+    }
+  }
+  return result.join("");
+}
+
+function bestGuess(candidates) {
+  const freq = Array.from({ length: 5 }, () => ({}));
+  for (const w of candidates) {
+    for (let i = 0; i < 5; i++) {
+      const c = w[i];
+      freq[i][c] = (freq[i][c] ?? 0) + 1;
+    }
+  }
+  let best = candidates[0];
+  let bestScore = -1;
+  for (const w of candidates) {
+    let scoreSum = 0;
+    for (let i = 0; i < 5; i++) {
+      scoreSum += freq[i][w[i]] ?? 0;
+    }
+    if (scoreSum > bestScore) {
+      bestScore = scoreSum;
+      best = w;
+    }
+  }
+  return best;
+}
+
+function loadState() {
+  if (!fs.existsSync(STATE_PATH)) return { candidates: words };
+  return JSON.parse(fs.readFileSync(STATE_PATH, "utf8"));
+}
+
+function saveState(state) {
+  fs.writeFileSync(STATE_PATH, JSON.stringify(state, null, 2));
+}
+
+const args = new Map(
+  process.argv.slice(2).map((arg) => {
+    const [k, v] = arg.replace(/^--/, "").split("=");
+    return [k, v];
+  }),
+);
+
+const guess = args.get("guess");
+const pattern = args.get("pattern");
+
+const state = loadState();
+if (guess && pattern) {
+  state.candidates = state.candidates.filter((w) => score(guess, w) === pattern);
+}
+
+const suggestion = bestGuess(state.candidates);
+saveState(state);
+
+console.log(`remaining=${state.candidates.length}`);
+console.log(`suggested=${suggestion}`);
+```
+
+### Run flow
+
+```bash
+agent-browser --session wordle open https://www.nytimes.com/games/wordle/index.html --headed
+
+# Close intro modal / cookie banner
+agent-browser --session wordle find role button click --name "Play"
+agent-browser --session wordle find role button click --name "Continue"
+agent-browser --session wordle find role button click --name "Close"
+agent-browser --session wordle press Escape
+
+# If needed, inspect refs
+agent-browser --session wordle snapshot -i -c -d 4
+```
+
+Submit a guess:
+
+```bash
+node ~/nexus/home/projects/wordle-solver.mjs
+# example output: suggested=slate
+
+agent-browser --session wordle click "body"
+for c in s l a t e; do agent-browser --session wordle press "$c"; done
+agent-browser --session wordle press Enter
+agent-browser --session wordle wait 1200
+```
+
+Extract the last row feedback:
+
+```bash
+# Use the NYT eval helper to avoid shell-escaping issues
+agent-browser --session wordle eval "$(cat {baseDir}/docs/wordle-nyt-eval.js)" --json
+```
+
+Feed the pattern back into the solver and repeat:
+
+```bash
+node ~/nexus/home/projects/wordle-solver.mjs --guess=slate --pattern=bybgb
+```
+
+Stop when the pattern is `ggggg`, then:
+
+```bash
+agent-browser --session wordle screenshot /tmp/agent-browser-wordle.png
+agent-browser --session wordle close
+```
+
+Notes:
+- If `pattern` includes `?`, wait longer and re-run the eval.
+- If a guess is rejected, your word list is wrong; re-download the Wordle list.
+
+### Fast path (Claude Code + Haiku 4.5)
+
+Recommended for demos and LLM-in-loop runs. This uses the optimized runner flags
+that minimize wait time and block ads.
+
+```bash
+claude --model haiku --print --permission-mode bypassPermissions \
+  "Run: bash ~/nexus/home/projects/wordle-nyt-run.sh \
+  --session demo-haiku \
+  --timing end-to-end \
+  --block-ads \
+  --wait-ms 0 \
+  --key-delay-ms 0 \
+  --poll-delay-ms 50 \
+  --max-poll-tries 14 \
+  --no-screenshot"
+```
+
+Notes:
+- The speedup comes from the runner flags, not model reasoning.
+- For loop-only timing, change to `--timing guess-only`.
+
+### Fast benchmarking (no cheating)
+
+Use the automation script to compare models or tune speed:
+
+```bash
+bash ~/nexus/home/projects/wordle-nyt-run.sh \
+  --session bench-haiku \
+  --first-guess crane \
+  --timing end-to-end \
+  --block-ads \
+  --wait-ms 0 \
+  --key-delay-ms 0 \
+  --poll-delay-ms 50 \
+  --max-poll-tries 14 \
+  --no-screenshot
+```
+
+Notes:
+- `--timing guess-only` excludes page load and modal dismissal.
+- Lower `--wait-ms` and `--key-delay-ms` reduces latency.
+- Lower `--poll-delay-ms` reduces time between visual update and action.
+- `--block-ads` can reduce page load time (best-effort).
+- Use unique `--session` names per run to avoid Wordle state collisions.

package/skills/agent-browser/docs/wordle-nyt-eval.js

@@ -0,0 +1,32 @@
+(() => {
+  const root = document.querySelector("#wordle-app-game");
+  if (!root) return { error: "wordle-app-game not found" };
+
+  const board = root.querySelector('div[class*="Board-module_board__"]');
+  if (!board) return { error: "board not found" };
+
+  const rows = Array.from(board.querySelectorAll('div[class*="Row-module_row"]'));
+  const tilesForRow = (row) => Array.from(row.querySelectorAll("[data-state]"));
+  const isFilled = (tiles) =>
+    tiles.length === 5 &&
+    tiles.every((t) => ["absent", "present", "correct"].includes(t.getAttribute("data-state")));
+
+  const filled = rows.map(tilesForRow).filter(isFilled);
+  const last = filled[filled.length - 1] ?? [];
+
+  const letters = last
+    .map((t) => (t.textContent ?? "").trim())
+    .join("")
+    .toLowerCase();
+  const pattern = last
+    .map((t) => {
+      const ev = t.getAttribute("data-state");
+      if (ev === "correct") return "g";
+      if (ev === "present") return "y";
+      if (ev === "absent") return "b";
+      return "?";
+    })
+    .join("");
+
+  return { letters, pattern, filledRows: filled.length };
+})()

package/skills/aix/SKILL.md

@@ -0,0 +1,93 @@
+---
+name: aix
+description: AI session intelligence - search and analyze your AI conversation history from Cursor, Claude Code CLI, Codex, Claude Desktop, and OpenCode
+homepage: https://github.com/Napageneral/aix
+metadata: {"nexus":{"emoji":"🧠","os":["darwin","linux"],"requires":{"bins":["aix"]},"install":[{"id":"brew","kind":"brew","formula":"Napageneral/tap/aix","bins":["aix"],"label":"Install via Homebrew"},{"id":"go","kind":"shell","script":"go install github.com/Napageneral/aix/cmd/aix@latest","bins":["aix"],"label":"Install via Go"}]}}
+---
+
+# aix 🧠
+
+AI session intelligence - search and analyze your AI conversation history from multiple sources.
+
+## Docs
+
+- `docs/setup.md`
+- `docs/embeddings.md`
+- `docs/troubleshooting.md`
+- `references/sql.md`
+
+## Supported Sources
+
+| Source | Tool | Status |
+|--------|------|--------|
+| `cursor` | Cursor IDE | Full messages |
+| `claude-code` | Claude Code CLI (`~/.claude/projects/`) | Full messages |
+| `codex` | Anthropic Codex CLI (`~/.codex/sessions/`) | Full messages |
+| `opencode` | OpenCode | Full messages |
+| `claude` | Claude Desktop | Metadata only |
+
+## What it does
+
+- Reads directly from AI tool databases (Cursor, Claude Code CLI, Codex, Claude Desktop, OpenCode)
+- Exports raw sessions to `~/nexus/home/sessions/` for git-tracked durability
+- Imports into `aix.db` for analysis and querying
+- Tracks which AI model was used (claude-4.5-opus, gpt-5, etc.)
+- Extracts rich metadata: capabilities, linter errors, file references, code blocks
+- Supports semantic search via embeddings
+
+## Quick Start
+
+```bash
+aix init                              # Initialize config and database
+aix sync --all                        # Sync from ALL sources
+aix sync --source claude-code         # Or sync from specific source
+aix sessions                          # List all sessions
+aix stats                             # Show statistics with source breakdown
+```
+
+## Key Commands
+
+| Command | Description |
+|---------|-------------|
+| `aix sync --all` | Sync from all available sources |
+| `aix sync --source <src>` | Sync from cursor/claude-code/codex/claude/opencode |
+| `aix sessions [--source <s>] [--project <p>]` | List sessions |
+| `aix show <session-id>` | View session (partial ID ok) |
+| `aix db query <sql>` | Run SQL queries |
+| `aix embed` | Generate embeddings |
+| `aix search <query>` | Semantic search |
+
+## Useful Queries
+
+```sql
+-- Sessions by source
+SELECT source, COUNT(*) as sessions, SUM(message_count) as msgs 
+FROM sessions GROUP BY source ORDER BY sessions DESC;
+
+-- Model usage
+SELECT model, COUNT(*) as sessions FROM sessions 
+WHERE model IS NOT NULL GROUP BY model ORDER BY sessions DESC;
+
+-- Recent activity by source
+SELECT date(created_at/1000, 'unixepoch') as day, source, COUNT(*) 
+FROM sessions WHERE created_at > 0 GROUP BY day, source ORDER BY day DESC LIMIT 14;
+```
+
+## Data Flow
+
+```
+Cursor DB       ─┐
+Claude Code CLI ─┼─► ~/nexus/home/sessions/ (durable) ─► aix.db (analysis)
+Codex CLI       ─┤
+Claude Desktop  ─┤
+OpenCode        ─┘
+```
+
+## Requirements
+
+- At least one AI tool (Cursor, Claude Code CLI, Codex, Claude Desktop, or OpenCode)
+- `GEMINI_API_KEY` (optional, for semantic search)
+
+## Source
+
+Project: `~/nexus/home/projects/aix`

package/skills/aix/docs/embeddings.md

@@ -0,0 +1,40 @@
+# aix Embeddings and Semantic Search
+
+Semantic search uses Gemini embeddings and requires `GEMINI_API_KEY`.
+
+## Set GEMINI_API_KEY
+
+```bash
+export GEMINI_API_KEY="your-key-here"
+```
+
+With 1Password:
+
+```bash
+export GEMINI_API_KEY=$(op read "op://Personal/Gemini API Key/credential")
+```
+
+Persist in your shell profile if desired:
+
+```bash
+echo 'export GEMINI_API_KEY="op://Personal/Gemini API Key/credential"' >> ~/.zshrc
+source ~/.zshrc
+```
+
+## Generate embeddings
+
+```bash
+aix embed
+```
+
+## Search
+
+```bash
+aix search "authentication bug"
+aix search "how to fix async error"
+```
+
+## Performance notes
+
+- First embed run can take several minutes on large histories.
+- Subsequent runs only process new messages.

package/skills/aix/docs/setup.md

@@ -0,0 +1,58 @@
+# aix Setup
+
+`aix` aggregates AI sessions across tools like Cursor, Claude Code CLI, Codex, Claude Desktop, and OpenCode.
+
+## Prerequisites
+
+- At least one supported AI tool with existing sessions
+- `aix` installed
+
+## Install
+
+```bash
+# Homebrew (recommended)
+brew install Napageneral/tap/aix
+
+# Or via Go
+go install github.com/Napageneral/aix/cmd/aix@latest
+
+# Verify
+aix --version
+```
+
+## Initialize
+
+```bash
+aix init
+```
+
+This creates:
+- `~/.aix/config.yaml`
+- `~/.aix/aix.db`
+- `~/nexus/home/sessions/` (exported sessions)
+
+## Sync sessions
+
+```bash
+aix sync --all
+aix sessions
+aix stats
+```
+
+Sync a specific source:
+
+```bash
+aix sync --source cursor
+aix sync --source claude-code
+aix sync --source codex
+aix sync --source claude
+aix sync --source opencode
+```
+
+## Verify
+
+```bash
+aix sessions
+aix stats
+aix show <session-id>
+```

package/skills/aix/docs/troubleshooting.md

@@ -0,0 +1,41 @@
+# aix Troubleshooting
+
+## No sources found
+
+**Symptom:** `aix sync --all` finds 0 sessions.
+
+**Fix:**
+1. Confirm the tool has been used at least once.
+2. Re-run sync for a specific source.
+
+```bash
+aix sync --source cursor
+```
+
+## Permission denied or locked DB
+
+**Cause:** AI tool is running or files are locked.
+
+**Fix:**
+1. Close the AI tool (Cursor, Claude Desktop, etc.).
+2. Re-run sync.
+
+```bash
+aix sync --all
+```
+
+## GEMINI_API_KEY missing
+
+**Cause:** Semantic search requires a key.
+
+**Fix:**
+
+```bash
+export GEMINI_API_KEY=$(op read "op://Personal/Gemini API Key/credential")
+```
+
+## Embeddings are slow
+
+**Cause:** Large history or rate limits.
+
+**Fix:** Let it finish; subsequent runs are incremental.

package/skills/aix/references/sql.md

@@ -0,0 +1,48 @@
+# aix SQL Reference
+
+## Core tables
+
+- `sessions` - session metadata (id, source, model, timestamps)
+- `messages` - individual messages
+- `code_blocks` - extracted code blocks
+- `capabilities` - capabilities detected
+- `files` - referenced files
+- `errors` - linter errors
+- `embeddings` - vector embeddings
+
+## SQL examples
+
+```sql
+-- Total sessions per source
+SELECT source, COUNT(*) as sessions
+FROM sessions
+GROUP BY source;
+
+-- Model usage
+SELECT model, COUNT(*) as count
+FROM sessions
+WHERE model IS NOT NULL
+GROUP BY model
+ORDER BY count DESC;
+
+-- Recent activity (last 14 days)
+SELECT date(created_at/1000, 'unixepoch') as day,
+       source,
+       COUNT(*) as sessions
+FROM sessions
+WHERE created_at > (strftime('%s', 'now') - 1209600) * 1000
+GROUP BY day, source
+ORDER BY day DESC;
+
+-- Sessions with specific file references
+SELECT s.id, s.title, f.path
+FROM sessions s
+JOIN files f ON s.id = f.session_id
+WHERE f.path LIKE '%auth%';
+
+-- Code blocks by language
+SELECT language, COUNT(*) as blocks
+FROM code_blocks
+GROUP BY language
+ORDER BY blocks DESC;
+```

package/skills/apple-notes/SKILL.md

@@ -0,0 +1,50 @@
+---
+name: apple-notes
+description: Manage Apple Notes via the `memo` CLI on macOS (create, view, edit, delete, search, move, and export notes). Use when a user asks Nexus to add a note, list notes, search notes, or manage note folders.
+homepage: https://github.com/antoniorodr/memo
+metadata: {"nexus":{"emoji":"📝","os":["darwin"],"requires":{"bins":["memo"]},"install":[{"id":"brew","kind":"brew","formula":"antoniorodr/memo/memo","bins":["memo"],"label":"Install memo via Homebrew"}]}}
+---
+
+# Apple Notes CLI
+
+Use `memo notes` to manage Apple Notes directly from the terminal. Create, view, edit, delete, search, move notes between folders, and export to HTML/Markdown.
+
+Setup
+- Install (Homebrew): `brew tap antoniorodr/memo && brew install antoniorodr/memo/memo`
+- Manual (pip): `pip install .` (after cloning the repo)
+- macOS-only; if prompted, grant Automation access to Notes.app.
+
+View Notes
+- List all notes: `memo notes`
+- Filter by folder: `memo notes -f "Folder Name"`
+- Search notes (fuzzy): `memo notes -s "query"`
+
+Create Notes
+- Add a new note: `memo notes -a`
+  - Opens an interactive editor to compose the note.
+- Quick add with title: `memo notes -a "Note Title"`
+
+Edit Notes
+- Edit existing note: `memo notes -e`
+  - Interactive selection of note to edit.
+
+Delete Notes
+- Delete a note: `memo notes -d`
+  - Interactive selection of note to delete.
+
+Move Notes
+- Move note to folder: `memo notes -m`
+  - Interactive selection of note and destination folder.
+
+Export Notes
+- Export to HTML/Markdown: `memo notes -ex`
+  - Exports selected note; uses Mistune for markdown processing.
+
+Limitations
+- Cannot edit notes containing images or attachments.
+- Interactive prompts may require terminal access.
+
+Notes
+- macOS-only.
+- Requires Apple Notes.app to be accessible.
+- For automation, grant permissions in System Settings > Privacy & Security > Automation.

package/skills/apple-reminders/SKILL.md

@@ -0,0 +1,67 @@
+---
+name: apple-reminders
+description: Manage Apple Reminders via the `remindctl` CLI on macOS (list, add, edit, complete, delete). Supports lists, date filters, and JSON/plain output.
+homepage: https://github.com/steipete/remindctl
+metadata: {"nexus":{"emoji":"⏰","os":["darwin"],"requires":{"bins":["remindctl"]},"install":[{"id":"brew","kind":"brew","formula":"steipete/tap/remindctl","bins":["remindctl"],"label":"Install remindctl via Homebrew"}]}}
+---
+
+# Apple Reminders CLI (remindctl)
+
+Use `remindctl` to manage Apple Reminders directly from the terminal. It supports list filtering, date-based views, and scripting output.
+
+Setup
+- Install (Homebrew): `brew install steipete/tap/remindctl`
+- From source: `pnpm install && pnpm build` (binary at `./bin/remindctl`)
+- macOS-only; grant Reminders permission when prompted.
+
+Permissions
+- Check status: `remindctl status`
+- Request access: `remindctl authorize`
+
+View Reminders
+- Default (today): `remindctl`
+- Today: `remindctl today`
+- Tomorrow: `remindctl tomorrow`
+- Week: `remindctl week`
+- Overdue: `remindctl overdue`
+- Upcoming: `remindctl upcoming`
+- Completed: `remindctl completed`
+- All: `remindctl all`
+- Specific date: `remindctl 2026-01-04`
+
+Manage Lists
+- List all lists: `remindctl list`
+- Show list: `remindctl list Work`
+- Create list: `remindctl list Projects --create`
+- Rename list: `remindctl list Work --rename Office`
+- Delete list: `remindctl list Work --delete`
+
+Create Reminders
+- Quick add: `remindctl add "Buy milk"`
+- With list + due: `remindctl add --title "Call mom" --list Personal --due tomorrow`
+
+Edit Reminders
+- Edit title/due: `remindctl edit 1 --title "New title" --due 2026-01-04`
+
+Complete Reminders
+- Complete by id: `remindctl complete 1 2 3`
+
+Delete Reminders
+- Delete by id: `remindctl delete 4A83 --force`
+
+Output Formats
+- JSON (scripting): `remindctl today --json`
+- Plain TSV: `remindctl today --plain`
+- Counts only: `remindctl today --quiet`
+
+Date Formats
+Accepted by `--due` and date filters:
+- `today`, `tomorrow`, `yesterday`
+- `YYYY-MM-DD`
+- `YYYY-MM-DD HH:mm`
+- ISO 8601 (`2026-01-04T12:34:56Z`)
+
+Notes
+- macOS-only.
+- If access is denied, enable Terminal/remindctl in System Settings → Privacy & Security → Reminders.
+- If running over SSH, grant access on the Mac that runs the command.