Skip to content

Module 1 — Individual-agent patterns

12 patterns diagnosing single-agent behavior. Discovery:

  • vstack-mcp list-tools — every tool name + friendly label
  • vstack-gbrain search "<your situation>" — semantic search (falls back to keyword scan if gbrain isn't installed)
  • The MCP resource vstack://patterns/index — live catalogue with summaries

The 12 patterns:

  1. #01 Lewin — vstack_lewin — attribution: INTERNAL (model) vs ENVIRONMENTAL (prompt/tools) vs INTERACTIONAL.
  2. #02 Goleman EI — vstack_goleman_ei — emotional-intelligence quadrants applied to agent dialogue.
  3. #03 Johari — vstack_johari — OPEN/BLIND/HIDDEN/UNKNOWN quadrants for agent self-awareness.
  4. #04 DANVA — vstack_danva_emotion — emotion-recognition cascade on user cues.
  5. #05 Cognitive reappraisal — vstack_cognitive_reappraisal — Gross's emotion regulation.
  6. #06 Yerkes-Dodson — vstack_yerkes_dodson — inverted-U workload curve.
  7. #07 HEXACO — vstack_hexaco — six-factor personality profile + drift detection.
  8. #08 Grant strengths — vstack_grant_strengths — strength-overuse pattern.
  9. #09 Motivation traps — vstack_motivation_traps — Saxberg-style motivation drift.
  10. #10 SDT — vstack_sdt_reward — Deci-Ryan self-determination theory.
  11. #11 McGregor — vstack_mcgregor — Theory X vs Theory Y on orchestrators.
  12. #12 Vroom — vstack_vroom_expectancy — expectancy theory (effort × instrumentality × valence).

See PATTERNS.md at the repo root for full per-pattern detail including citations.