## Intel — 2026-01-08 (ElizaOS)

### 1) Data Pattern Recognition

**Development velocity (last 48h, Jan 7–8)**
- GitHub (elizaos/eliza): **2 PRs opened, 1 merged; 7 issues opened; 7 active contributors**
- Change theme concentration:
  - **Connector compatibility regression** (core v1.7.0 ↔ plugin-discord v1.3.3) dominating support load
  - **Database portability**: PGLite-specific fixes surfaced (pgcrypto extension handling)
  - **Infrastructure correctness**: TOCTOU race-condition fixes and runtime init optimization in progress

**Trend signal**
- **Regression-driven throughput:** Engineering effort is being pulled toward “fix-forward” on v1.7.0 rather than net-new feature velocity.
- **Cross-repo coupling is the risk amplifier:** Core schema/migration changes (“serverId → messageServerId”) are cascading into plugin-bootstrap, plugin-discord, and SQL schema references.

**Community engagement patterns (Discord)**
- High-intensity support threads cluster around:
  - **Discord connector breakage** (“No server ID found 10”)
  - **Cloud API configuration confusion** (“Model not found” due to model naming format)
  - **Token/CA discoverability** (trust + onboarding friction)
- The fastest resolutions occurred when guidance was **copy/paste actionable** (e.g., exact model string formats), indicating the community benefits disproportionately from “small, precise docs.”

---

### 2) User Experience Intelligence

**Feedback categorized by impact × theme**

**P0 — Platform breakage / adoption blocker**
- **Discord bot integration broken on core v1.7.0**
  - Symptom: “No server ID found 10” even with correct intents/admin perms
  - Root cause: incomplete migration from `serverId` to `messageServerId` across bootstrap actions/providers + compatibility mismatch with plugin-discord **v1.3.3**
  - Mitigations offered: downgrade core to **v1.6.5** or use fix branch **odi-17**
  - User impact: prevents Discord onboarding for new agents; causes repeated permission misdiagnosis and wasted setup time

**P1 — Configuration/documentation friction**
- **Cloud agent API “Model not found”**
  - Resolved via correct `provider/model` format:
    - `openai/gpt-4o-mini`, `anthropic/claude-sonnet-4.5`, `google/gemini-2.5-flash`
  - Pattern: users assume model is a bare name; system requires provider prefix; error message is not self-correcting

**P1 — Trust/onboarding friction (non-technical)**
- **Contract address (CA) discoverability**
  - Users report inability to find official CA quickly on X/Linktree
  - Team intent: refresh Linktree → CoinGecko; post CA more visibly (with concern that “CA in bio” can look scammy)
  - UX risk: trust gap during onboarding; higher support burden; increased scam surface area

**P2 — UI ergonomics**
- Issue #6310: chat box default sizing should start at **1 line** and expand dynamically
  - Indicates ongoing polish needs; likely improves perceived quality but not a blocker

**Usage vs intended design (observed)**
- Users are debugging Discord failures as **permissions/intents problems**, but failures are **schema/migration incompatibilities**. This mismatch implies we need:
  - Better runtime diagnostics (“expected messageServerId, got undefined; check versions X/Y”)
  - Stronger version compatibility signaling

---

### 3) Strategic Prioritization (Impact × Risk × Dependencies)

#### A) Stabilize Discord integration on core v1.7.0 (P0)
**What’s known**
- PR **#6333** (merged) updates plugin-bootstrap (+ SQL minor) actions/providers for `serverId → messageServerId`.
- Still requires **validation against plugin-discord v1.3.3** and “various Discord branches,” plus likely a **new plugin-discord release** after testing.

**Dependencies / critical path**
1. Confirm core release artifact includes #6333 (and any companion fixes)
2. Compatibility matrix test: core **1.7.0+** × plugin-discord **1.3.3** (and next patch)
3. Cut patched core release (if not already) + plugin-discord release (if required)
4. Publish upgrade/downgrade guidance with exact version pairs

**Recommendation (resource allocation)**
- Allocate **1 owner (core)** + **1 owner (plugin-discord)** for a 24–48h stabilization sprint:
  - Create a minimal reproducible harness (Discord test guild + scripted flows)
  - Add a CI smoke test that asserts server ID resolution end-to-end
- Ship: **core patch release** (if needed) + **plugin-discord patch** + pinned Discord announcement.

**Success metric**
- Reduce Discord integration support threads from “recurring daily” to **near-zero within 72h** after release + announcement.
- Add a “known-good versions” table; measure drop in repeated questions.

#### B) Make model configuration errors self-healing (P1)
**Observed fix**
- Users succeed immediately when given correct model strings.

**Recommendation**
- Productize the fix in two places:
  1. **API validation:** if `model` lacks a `/`, return an error like:  
     “Model must be `provider/model` (e.g., `openai/gpt-4o-mini`). You sent: `gpt-4o-mini`.”
  2. **Docs quickstart + SDK examples:** ensure every snippet uses `provider/model`.

**Success metric**
- Track “Model not found” reports in Discord: target **>80% reduction** after improved error messaging + docs.

#### C) Reduce cross-repo migration blast radius (P1 / technical debt containment)
**Problem pattern**
- Schema rename (`serverId → messageServerId`) created a multi-surface regression across bootstrap + discord + SQL schema + user runtime behavior.

**Recommendation**
- Establish a **Migration Completion Gate** for breaking field renames:
  - One checklist spanning: core types, plugin-bootstrap, plugin-* connectors, DB schema, docs, and example configs.
  - Require compatibility tests before tagging a release that includes renamed fields.
- Add a short-lived compatibility shim:
  - If safe, accept both fields at runtime; log deprecation warnings with a link to upgrade instructions.

**Success metric**
- Fewer “version mismatch” regressions: target **0 P0 regressions** caused by partial migrations in the next 30 days.

#### D) Trust & discoverability: CA visibility without “scammy” presentation (P1)
**Recommendation**
- Implement a **single canonical CA page** on the official website and link it everywhere:
  - X pinned post → canonical page
  - Linktree → CoinGecko + canonical page
  - Docs “Security / Official Links” section
- Add “how to verify” steps (chain explorer link, checksum, warning about impersonators).

**Success metric**
- Reduce repeated CA questions; improve onboarding confidence (qualitative sentiment shift; fewer “can’t find CA” messages).

#### E) Architecture scaling: event pumps + voice/text prioritization (P2 now, P1 soon)
**Current direction**
- “Simple event pumps” + multiple daemon instances per service; voice needs priority/bandwidth; preprocessing helps.

**Recommendation**
- Convert discussion into an executable RFC milestone:
  - Define QoS tiers (voice vs text), backpressure strategy, and daemon horizontal scaling model.
  - Identify integration point with existing Discord gateway work (Jeju branch as reference).

**Success metric**
- RFC acceptance + implementation plan with measurable throughput/latency targets (e.g., queue latency, drop rates under load).

---

### Actionable Next Steps (Next 24–72 hours)

1. **Release engineering**
   - Verify #6333 is included in the next core patch; if not, cut **core v1.7.1** (or equivalent).
   - Run compatibility tests against plugin-discord **v1.3.3**; if failing, prioritize plugin-discord patch + release.

2. **Diagnostics & messaging**
   - Improve runtime error for missing server ID to include: detected core/plugin versions + expected field (`messageServerId`) + link to compatibility table.

3. **Docs micro-fixes with high ROI**
   - Add “Model parameter format” to the top of the Cloud API quickstart and return better API validation errors.
   - Publish a pinned Discord message: “Known-good version pairs + downgrade instructions.”

4. **Trust surface**
   - Create canonical CA verification page + pin on X; update Linktree to CoinGecko and that canonical page.