# ElizaOS Intel — 2026-01-26

## 1) DATA PATTERN RECOGNITION

### Development velocity & trend (GitHub, month-to-date snapshot)
- **PRs:** 36 opened / **20 merged** (≈ **55.6% merge rate**)
- **Issues:** 49 opened / **42 closed** (≈ **85.7% closure rate**)
- **Contributors:** **28 active**
- **Code churn:** **+25,257 / -8,657** across **261 files** (high-change month; elevated integration risk)

**Trend signal:** Core + plugin reliability work is landing quickly (high issue closure), while **V2.0.0 workstreams (Rust/WASM + dynamic execution engine + Python bridge)** are large and still in-flight → risk of attention fragmentation and QA bottlenecks.

### Community engagement patterns (Discord, last 3 days)
- Discussion concentrates into **3 recurring lanes**:
  1) **MCP/Eliza Cloud integration** (external MCP registration, proxying)
  2) **Operational reliability** (CLI version conflicts, plugin errors, self-hosting)
  3) **Token/economics governance** (ecosystem tokens, legitimacy, incentives)
- **Unanswered recurring questions (high visibility):**
  - Token relationships/impact on core project (**asked at least twice**: Jan 25 + broader context from Jan 23–24)
  - Why more “single-purpose agents” aren’t visible in the wild (**asked Jan 25**)

### Feature adoption & usage signals
- **External MCP for Cloud agents is a real demand driver** (AI red teaming + human-in-loop). Concrete endpoint shared: `POST /api/v1/mcps`.
- **Self-hosting is trending toward low-cost, always-on setups**:
  - Mini-PC deployments cited as a community pattern
  - Oracle Cloud Free Tier guide referenced (**4 vCPU / 24GB RAM**) → indicates demand for “official blessed infra recipes”.

### Pain point correlation across channels
- **Reliability pain points correlate directly with adoption friction**:
  - CLI update/migration conflicts (1.6.5 → 1.7.2), cached versions, SQL migrations
  - Discord plugin errors resolved via version bump (plugin-discord **1.3.8**), but **DM edge cases persist** (“User has no name or username”)
- **Security incidents are repeating**:
  - Scam “Create A Ticket” bot requesting wallet addresses (Jan 25)
  - Token “manual migration” scam with wallet address & urgency framing (Jan 24)
  - Pattern suggests automated targeting when users ask support questions publicly.

---

## 2) USER EXPERIENCE INTELLIGENCE

### Feedback themes (categorized by impact)

**P0 — Trust & Safety / Brand Risk**
- Multiple active scams leveraging “support” framing (wallet requests, suspicious links).
- Community uncertainty about which tokens are official/related increases scam susceptibility.

**P1 — Onboarding & Adoption Friction**
- “Why aren’t there more one-trick agents?” indicates a gap between framework capability and perceived ease-to-ship.
- External MCP registration is powerful but currently “insider knowledge” (endpoint shared in chat; docs action item exists).

**P1 — Upgrade & Compatibility Reliability**
- CLI upgrade issues required multi-step manual remediation (cache clears, lockfile deletion).
- Discord plugin fixes exist but DM reliability remains unresolved; DM is a core onboarding/support surface.

**P2 — Infra Enablement**
- Strong interest in **cheap 24/7 hosting** (mini PC, Oracle free tier). Opportunity for official guides + reference images.

### Usage patterns vs intended design (mismatches)
- Users are using Discord as **primary support + release notes** → when answers are missing (token relationships, best-practice agent patterns), uncertainty becomes persistent narrative.
- Developers are seeking **Cloud-first extensibility** (external MCP) rather than only local plugin development → docs and UI likely lagging this reality.

### Sentiment tracking (directional)
- Technical sentiment: constructive, pragmatic (MCP integration help, version fix steps).
- Governance/market sentiment: strained (token launch controversy; mention of limited “green days”; calls for buybacks; founder frustration). This is amplifying scrutiny of anything token-adjacent.

---

## 3) STRATEGIC PRIORITIZATION (Impact × Risk × Dependency)

### Priority A — Trust, clarity, and scam suppression (High impact, low-to-medium engineering risk)
**Why now:** Repeated scams + unclear token relationships are compounding into trust loss and support load.

**Actions (next 72 hours)**
1) **Pin an “Official Support & Never Ask” policy** in Discord:
   - “No staff will ever request wallet address / token transfer”
   - Official migration links and verification steps
2) **Token relationship explainer (single source of truth)**:
   - “Core token vs ecosystem tokens vs unofficial/community tokens”
   - Include: who can claim “official”, what endorsements mean, and risk disclaimers
3) **Automod hardening**:
   - Block common scam phrases (“send tokens”, “manual migration”, “ticket wallet”)
   - Restrict newly-added bots; require admin approval + audit

**Success metric targets**
- Reduce scam reports to **≤1/week**
- Increase answered “token relationship” questions to **100%** via linkable doc

---

### Priority B — MCP as a flagship developer experience (High impact, medium risk, depends on docs + product surface)
**Why now:** External MCP is emerging as the clearest “aha” capability (red teaming, tool use, proxy execution). Current knowledge is chat-level, not product-level.

**Actions (next 1–2 sprints)**
1) **Publish “External MCP in Eliza Cloud” guide**:
   - Endpoint: `POST /api/v1/mcps`
   - Minimal example payload + auth model + expected lifecycle
   - Security considerations (allowlists, rate limits, request logging)
2) **Add a Cloud UI affordance** (even minimal):
   - “Add MCP Server” form that calls the same endpoint
3) **Ship 2 reference MCP servers**:
   - “Red Teaming checklist + evidence capture”
   - “Video editing MCP (DaVinci) capabilities + limitations documented”

**Success metric targets**
- **2+** community-built MCP servers registered/week
- Time-to-first-tool-call for new devs **<30 minutes** (from docs)

---

### Priority C — Reduce upgrade friction (High impact, medium risk, cross-repo dependency)
**Why now:** Version conflicts and migrations are blocking upgrades and generating repeated support threads.

**Actions (next sprint)**
1) **CLI “doctor” command**:
   - Detect cached CLI mismatch, lockfile drift, plugin/core version incompatibility
   - Offer auto-fix steps (or print exact commands)
2) **Compatibility matrix** pinned in docs:
   - Core version ↔ plugin-discord version (e.g., **1.7.2 ↔ 1.3.8**)
3) **DM reliability fix for Discord plugin**:
   - Reproduce “User has no name or username” path
   - Add fallback identifiers + structured logging (avoid silent skipping)

**Success metric targets**
- Cut “update stuck on old version” incidents by **50%**
- DM error rate: measurable and trending down (instrumentation first)

---

### Priority D — Enable “one-trick pony” agent adoption (High impact, higher product risk, requires packaging + distribution)
**Why now:** The question “why no specialized agents?” signals packaging/distribution gaps, not just missing features.

**Actions (2–4 weeks)**
1) **Template gallery**: 5 minimal single-purpose agents
   - “Prediction agent” (Sportradar plugin)
   - “Support triage agent”
   - “Content clipper agent” (transcription trimming)
   - “Monitoring/alerts agent”
   - “Red teaming agent” (external MCP)
2) **“Deploy in 1 hour” paths**:
   - Oracle free tier reference deployment
   - Mini-PC guide (bill of materials + power/network notes)
3) **Incentivize public sharing**:
   - Tie to existing “public agent discovery” direction: standard URLs + discovery module

**Success metric targets**
- **+5** publicly listed specialized agents/week
- Template-to-deploy conversion (downloads → active agents) tracked

---

## Resource allocation recommendation (next 2 sprints)
- **30%** Trust & Safety + comms (docs, pinned policies, automod, token clarity)
- **30%** MCP developer experience (docs + minimal UI + reference servers)
- **25%** Reliability/upgrade friction (CLI doctor, plugin DM fix, version matrix)
- **15%** Adoption packaging (one-trick templates + deploy recipes)

**Key dependency callout:** Token ecosystem ambiguity is currently a “force multiplier” for both scam effectiveness and community conflict. Addressing it will reduce support load and improve receptiveness to technical launches (Cloud apps, Hyperscape MVP, V2.0.0 progress).