Update to Basic Memory Cloud

{
  "name": "memory-reflect",
  "description": "Sleep-time memory reflection: review recent conversations and daily notes, extract insights, and consolidate into long-term memory. Use when triggered by cron, heartbeat, or explicit request to reflect on recent activity. Runs as background processing to improve memory quality over time.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 385
    }
  ],
  "skill_md_contents": "---\nname: memory-reflect\ndescription: \"Sleep-time memory reflection: review recent conversations and daily notes, extract insights, and consolidate into long-term memory. Use when triggered by cron, heartbeat, or explicit request to reflect on recent activity. Runs as background processing to improve memory quality over time.\"\n---\n\n# Memory Reflect\n\nReview recent activity and consolidate valuable insights into long-term memory.\n\nInspired by sleep-time compute — the idea that memory formation happens best *between* active sessions, not during them.\n\n## When to Run\n\n- **Cron/heartbeat**: Schedule as a periodic background task (recommended: 1-2x daily)\n- **On demand**: User asks to reflect, consolidate, or review recent memory\n- **Post-compaction**: After context window compaction events\n\n## Process\n\n### 1. Gather Recent Material\n\nFind what changed recently, then read the relevant files:\n\n```python\n# Find recently modified notes — use json format for the complete list\n# (text format truncates to ~5 items in the summary)\nrecent_activity(timeframe=\"2d\", output_format=\"json\")\n\n# Read specific daily notes\nread_note(identifier=\"memory/2026-02-27\")\nread_note(identifier=\"memory/2026-02-26\")\n\n# Check active tasks\nsearch_notes(note_types=[\"task\"], status=\"active\")\n```\n\n### 2. Evaluate What Matters\n\nFor each piece of information, ask:\n- Is this a **decision** that affects future work? → Keep\n- Is this a **lesson learned** or mistake to avoid? → Keep\n- Is this a **preference** or working style insight? → Keep\n- Is this a **relationship** detail (who does what, contact info)? → Keep\n- Is this **transient** (weather checked, heartbeat ran, routine task)? → Skip\n- Is this **already captured** in MEMORY.md or another long-term file? → Skip\n\n### 3. Update Long-Term Memory\n\nWrite consolidated insights to `MEMORY.md` following its existing structure:\n- Add new sections or update existing ones\n- Use concise, factual language\n- Include dates for temporal context\n- Remove or update outdated entries that the new information supersedes\n\n### 4. Log the Reflection\n\nAppend a brief entry to today's daily note:\n```markdown\n## Reflection (HH:MM)\n- Reviewed: [list of files reviewed]\n- Added to MEMORY.md: [brief summary of what was consolidated]\n- Removed/updated: [anything cleaned up]\n```\n\n## Guidelines\n\n- **Be selective.** The goal is distillation, not duplication. MEMORY.md should be curated wisdom, not a copy of daily notes.\n- **Preserve voice.** If the agent has a personality/soul file, reflections should match that voice.\n- **Don't delete daily notes.** They're the raw record. Reflection extracts from them; it doesn't replace them.\n- **Merge, don't append.** If MEMORY.md already has a section about a topic, update it in place rather than adding a duplicate entry.\n- **Flag uncertainty.** If something seems important but you're not sure, add it with a note like \"(needs confirmation)\" rather than skipping it entirely.\n- **Restructure over time.** If MEMORY.md is a chronological dump, restructure it into topical sections during reflection. Curated knowledge > raw logs.\n- **Check for filesystem issues.** Look for recursive nesting (memory/memory/memory/...), orphaned files, or bloat while gathering material.\n"
}