drift

name	drift
description	Stage N1. Stochastic scanning of past conversations to gather raw memory fragments. Broad, unfocused, deliberately noisy. The only stage that touches conversation history directly.
context	fork
user-invocable	false
allowed-tools	["Read","Grep","Glob","Bash"]

Drift — Stage N1

The hypnagogic scan. Fast, cheap, broad, noisy. ~5% of cycle time.

Drift is the only stage that directly searches past conversations. All downstream stages work exclusively with drift's output.

Inputs

From orchestrator: target_skill, scan_depth (default 10), recency_bias (0-1, default 0.7), randomness (0-1, default 0.4), cycle_number (1-indexed).

Procedure

1. Extract Search Terms

Read the target SKILL.md. Extract:

• Skill name and description keywords
• File types it handles (.docx, .xlsx, etc.)
• Domain terms (presentation, spreadsheet, formatting)
• Action verbs specific to the skill

Build a keyword pool of 10-20 terms ranked by specificity.

2. Sample Conversations

Use conversation_search and recent_chats tools. Strategy varies by cycle:

Cycle	Recency Bias	Randomness	Scan Depth
1	0.8	0.3	8
2	0.5	0.5	10
3	0.3	0.7	12
4+	0.2	0.8	15

Early cycles: mostly recent_chats. Later: mostly conversation_search with random keywords from the less-specific end of the pool. Include cross-domain terms in later cycles.

3. Extract Fragments

From each result, extract 1-3 sentence fragments capturing:

• What the user asked for
• What tools/skills were involved
• Friction signals (corrections, rephrasing, frustration)
• Unusual elements (unexpected file types, edge requirements)

Do NOT reproduce conversation content — summarize the essence.

Fragment format (YAML):

- id: f001
  source: recent_chats | conversation_search
  recency: recent | mid | old
  relevance: direct | adjacent | tangential
  content: "User requested Word doc with mixed English/Arabic, bidirectional TOC"
  friction: true
  friction_signal: "Corrected RTL formatting twice"
  tags: [docx, rtl, toc, multilingual]
  synthetic: false

4. Inject Noise

Add 1-3 synthetic fragments — deliberately tangential or surreal:

• Domain mashup: Combine skill domain with random unrelated domain
• Scale distortion: Warp a quantity from a real fragment to extremes
• Inversion: Flip an assumption from a real fragment

Tag all as synthetic: true.

5. Output Fragment Bag

fragment_bag:
  target_skill: /path/to/SKILL.md
  cycle_number: 1
  total_fragments: 14
  sources: {recent_chats: 7, conversation_search: 4, synthetic: 3}
  fragments: [...]

Behavioral Rules

• Be fast. Don't over-analyze. If vaguely relevant, include it.
• Be broad. Include tangential material. Cross-pollination needs it.
• Be noisy. Synthetic fragments are not optional.
• Don't judge. Filtering is N2's job, not yours.

Edge Cases

• No conversations found: Generate 5-8 fully synthetic fragments from the skill's documented scope. Dreams don't require real memories.
• Very few results (<3): Supplement with synthetic to minimum bag of 5.