@aman_asmuei/aman-agent 0.25.0 → 0.27.0

package/README.md

@@ -47,7 +47,7 @@
 <details>
 <summary><strong>Table of Contents</strong></summary>
 
-- [What's New](#whats-new-in-v0240)
+- [What's New](#whats-new-in-v0270)
 - [The Problem](#the-problem)
 - [The Solution](#the-solution)
 - [Quick Start](#quick-start)
@@ -80,26 +80,38 @@
 
 ---
 
-## What's New in v0.24.0
+## What's New in v0.27.0
 
-> **Session telemetry, post-mortem analysis, and a smarter feedback loop.**
+> **Your agent remembers you across sessions.**
 
-The agent now passively observes what happens during a session — tool calls, file changes, sentiment shifts, blockers, milestones — and on session end can generate a structured post-mortem with actionable patterns stored back as memories.
+Every session now contributes to a persistent **dynamic user model** — trust score, sentiment baseline, energy patterns, frustration correlations — all computed from real session signals. The next session starts calibrated to *your* actual patterns instead of generic time-based heuristics. Zero new LLM calls.
 
 | Feature | What it does |
 |:---|:---|
-| **Observation system** | Passive event capture: tool calls, errors, file changes, sentiment, topic shifts, blockers, milestones — all written to JSONL, never blocking the user |
-| **Post-mortem reports** | LLM-generated session analysis with goals, completed work, blockers, decisions, sentiment arc, recurring patterns, and recommendations |
-| **Smart auto-trigger** | Reports generate automatically when sessions hit ≥3 tool errors, ≥2 blockers, &gt;60min, abandoned plan steps, or sustained frustration |
-| **Pattern memory loop** | Recurring patterns from post-mortems are stored as `pattern` memories — your AI learns from its own session history |
-| **`/observe` dashboard** | Live session stats: tool calls, errors, files changed, blockers, milestones, topic shifts |
-| **`/postmortem` commands** | Generate now, view last, list all, or `--since 7d` for cross-session trend analysis |
-| **Eval reliability fix** | `/eval report` now uses `fs.existsSync` instead of an ANSI-sensitive prefix check |
-| **CI dependency stability** | Switched `npm ci` → `npm install` to handle platform-specific native deps |
+| **Dynamic user model** | Cross-session profile: trust (EMA), sentiment baseline, energy distribution, engagement trends |
+| **Feed-forward** | Next session starts with personalized energy/mode overrides based on your history |
+| **Trust scoring** | 0–100% trust from explicit ratings + implicit signals (frustration, blockers, milestones) |
+| **Frustration correlations** | Pearson r analysis: are tool errors, long sessions, or late nights causing frustration? |
+| **Nudge tracking** | Tracks which wellbeing nudges fire and session quality after — data for adaptive nudging |
+| **`/identity dynamics`** | View your user model: trust, sentiment trend, preferred time, avg session length |
+| **`/identity dynamics --json`** | Raw JSON dump of the full user model |
+| **`/identity dynamics --reset`** | Start fresh — delete the model and rebuild |
+| **acore Dynamics sync** | Trust %, session count, and sentiment trend written to `## Dynamics` in core.md |
 
 <details>
 <summary><strong>Highlights from earlier releases</strong></summary>
 
+**v0.26 — Skill crystallization**
+- Post-mortems identify reusable procedures → opt-in prompt → saved as auto-triggering skills
+- Runtime trigger matching, `/skills crystallize`, `/skills list --auto`
+- Post-mortem JSON sidecar for lossless re-parsing
+
+**v0.24 — Observation & post-mortem**
+- Passive session telemetry (tool calls, errors, file changes, sentiment, blockers)
+- LLM-generated post-mortem reports with smart auto-trigger
+- Pattern memory loop — the agent learns from its own session history
+- `/observe` dashboard, `/postmortem` commands
+
 **v0.18 — User onboarding**
 - Interactive first-run setup capturing name, role, expertise, communication style
 - 13 showcase templates (fitness, freelancer, Muslim, finance, etc.)
@@ -1040,11 +1052,11 @@ On session end, if any smart trigger fires (≥3 tool errors, ≥2 blockers, &gt
 | `/profile` | Your profile + agent profiles `[me\|edit\|setup\|create\|list\|show\|delete]` |
 | `/delegate` | Delegate task to a profile `[<profile> <task>\|pipeline]` |
 | `/team` | Manage agent teams `[create\|run\|list\|show\|delete]` |
-| `/identity` | View identity `[update <section>]` |
+| `/identity` | View identity `[update <section>]` `[dynamics [--json\|--reset]]` |
 | `/rules` | View guardrails `[add\|remove\|toggle ...]` |
 | `/workflows` | View workflows `[add\|remove ...]` |
 | `/tools` | View tools `[add\|remove ...]` |
-| `/skills` | View skills `[install\|uninstall ...]` |
+| `/skills` | View skills `[install\|uninstall\|crystallize\|list --auto]` |
 | `/eval` | View evaluation `[milestone ...]` |
 | `/memory` | View memories `[search\|clear\|timeline]` |
 | `/observe` | Live session telemetry dashboard `[pause\|resume]` |