# ElizaOS Strategic Intel — 2026-01-30

## 1) Data Pattern Recognition (Velocity, Trends, Correlations)

### Development velocity (GitHub, Jan-to-date snapshot)
- **Monthly throughput (2026-01-01 → 2026-02-01 window, current)**: **38 PRs opened / 22 merged**, **96 issues opened / 55 closed**, **31 active contributors**.
- **Code churn**: **+25,706 / -8,753** across **269 files**, **379 commits** (high change volume consistent with v2 refactors + plugin infra work).
- **Execution trend**: strong emphasis on “make it run reliably” work (CLI bootstrap, SQL reliability, streaming/transport, env/secrets correctness) alongside **major v2.0.0 branch work** and **n8n automation pivot**.

**Interpretation**
- Velocity is healthy, but **risk is rising** due to large architectural deltas (v2 branch, workflow engine integration, routing ideas) while **user-facing breakages** persist (callbacks ordering, plugin install/module resolution, migration liquidity UX).

### Community engagement patterns (Discord last 72h)
Conversation load clusters into 4 themes:
1) **Token/migration economics** (high heat, negative sentiment, repeated questions)
2) **Workflow automation + integrations** (n8n, Composio, OAuth, connection page)
3) **Reliability/debuggability** (embeddings providers, SSE issues, Cursor AUTO pitfalls)
4) **Core framework correctness** (action callback ordering bug in 1.7.2)

**Correlation pattern**
- When **migration + token utility** questions spike, dev-support questions also spike (module resolution, bridging, liquidity). Community sees these as a single “trust & operability” surface, not separate topics.

### Feature adoption / usage vs intended design (signals)
- **Embeddings**: users frequently misconfigure `EMBEDDING_PROVIDER` (attempting `openrouter`) and hit validation; OpenAI is perceived as the only “reliably working” path. This indicates a **documentation + defaults problem**, not only provider issues.
- **Streaming transport**: SSE setup produces MIME type mismatch errors; community workaround is **“use socket.io instead of SSE”**, implying SSE path is fragile or poorly documented for common deployments.
- **Routing**: currently “alternate small/large models” is understood as insufficient; community is proactively ideating **router-based selection** (RouteLLM-like prompt complexity routing).

---

## 2) User Experience Intelligence (Themes, Impact, Sentiment, Opportunities)

### Feedback themes by impact

#### P0 — Trust blockers / financial UX failures
- **ETH chain swap/migration UX**: reports of **~98% loss** when swapping to ETH/USDT due to **zero liquidity**.
- **Token distribution controversy**: repeated questions about **40% team allocation** post 1:10 increase and **lack of disclosed vesting/wallets**.

**Sentiment**: strongly negative; risk of narrative capture (“open source but team got 40%”) and ongoing churn in public channels.

**Opportunity**
- Treat as **support + transparency incident**, not “community debate”. Provide authoritative, versioned answers and a single canonical migration path with warnings.

#### P0 — Framework correctness / developer productivity
- **Eliza 1.7.2 action callbacks bug**
  - Expected: (1) initial feedback → (2) structured return → (3) detailed callback message
  - Actual: reversed ordering; **structured return omitted**
  - Affects: custom plugins using **plugin-sql, plugin-openai, plugin-bootstrap**
  
**Impact**: breaks documented developer mental model; causes “agent feels broken” UX and plugin ecosystem regressions.

#### P1 — Onboarding friction / “it doesn’t install”
- `Cannot find module '@elizaos/plugin-web-search'` after install; workaround requires editing `package.json` / using bun.
- Cursor AUTO mode reported to “break apps” after long debugging; guidance is “use Composer 1”.

**Impact**: increased time-to-first-success; higher support load; undermines “minutes to start” goal.

#### P1 — Integration UX direction is converging (positive)
- **Connection page** plan (link social accounts → redirect back to bot) is aligned with standard OAuth UX.
- **Composio** discovery: already provides **in-chat auth** + 100+ integrations; plus an existing **plugin-composio** repo and forthcoming RFC.

**Opportunity**
- Fast-track Composio evaluation to reduce custom auth surface area and accelerate top-of-funnel integrations.

---

## 3) Strategic Prioritization (Impact vs Risk, Dependencies, Resource Allocation)

### Priority stack (next 7–14 days)

#### P0: Stabilize “developer trust surface”
1) **Fix Eliza 1.7.2 callback ordering + missing structured return**
   - **User impact**: very high (breaks plugins + conversational UX)
   - **Tech risk**: medium (ordering/async semantics; potential breaking changes)
   - **Definition of done**:
     - Callback sequence matches docs in deterministic order
     - Structured return always delivered
     - Add regression tests covering plugin-sql + plugin-openai + plugin-bootstrap interaction
2) **Migration “single source of truth” + ETH liquidity warning banner**
   - **User impact**: very high (prevents catastrophic user outcomes)
   - **Tech risk**: low–medium (mostly comms + routing users to correct path)
   - **Definition of done**:
     - One canonical migration guide (steps, chain-specific warnings, “do not swap on zero-liquidity pools”)
     - FAQ addressing 40% allocation question + vesting/wallet disclosure plan (even if partial with timeline)

**Resource recommendation**: allocate a named DRI for each (1 eng for callback bug; 1 ops/comms + 1 eng liaison for migration docs + site banner).

#### P1: Convert integration momentum into a shippable funnel
3) **Connection page + OAuth gateway decision (Composio vs internal)**
   - **Dependencies**: OAuth credential storage model, redirect URIs, multi-tenant security
   - **Recommendation**: run a **48-hour spike**:
     - Integrate Stan’s `plugin-composio` into a minimal “connect → redirect → action call” flow
     - Decide: Composio-first for breadth vs internal-first for control
4) **n8n workflow engine path (v2)**
   - **Critical path**: OAuth gateway + credential passing from plugins → n8n execution
   - **Recommendation**:
     - Ship with hosted n8n execution first; defer self-host optimization
     - Build a “prefab workflow library” + “experimental generated workflows” QA loop (as discussed)

**Resource recommendation**: 2 engineers for integrations (connection page + Composio spike), 1 engineer for n8n workflow library scaffolding, 1 engineer for execution/runtime plumbing.

#### P2: Cost/performance + routing (strategic but not blocking)
5) **Embeddings/provider UX hardening**
   - Fix defaults/docs so users don’t set `openrouter` as provider where invalid; clarify supported matrix.
   - Add a “known good minimal config” and explicitly document cost multiplier of contextual embeddings (`CTX_KNOWLEDGE_ENABLED=true`).
6) **Router exploration (“ElizaRouter”)**
   - **Recommendation**: keep as research until P0/P1 stabilized; define success metrics:
     - cost reduction %, latency improvement %, accuracy/quality delta
   - Start with a small deterministic router heuristic before training/hosting small models.

---

## Quantitative KPI Snapshot (from available signals)
- **Critical unresolved bugs surfaced**: 1 (callbacks ordering) with multi-plugin blast radius.
- **High-severity user loss reports**: 1 class (ETH chain zero-liquidity swap leading to ~98% loss).
- **Integration acceleration artifacts available**: 1 existing plugin (`plugin-composio`) + RFC pending.
- **Monthly engineering capacity signal**: 31 contributors / 22 merged PRs suggests bandwidth exists, but must be focused to reduce support drag.

---

## Actionable Recommendations (Immediate)
1) **Ship a patch + test** for callback ordering in 1.7.2 (or release 1.7.3) and publish a short “behavioral contract” doc for callbacks.
2) **Publish migration incident-grade guidance**: pinned Discord message + website banner + step-by-step doc; include explicit “do not swap on ETH if liquidity is zero” warning.
3) **Composio decision sprint**: integrate plugin-composio into the connection page MVP; decide within 1 week whether Composio becomes the default auth/integration substrate.
4) **Reduce onboarding entropy**: turn recurring fixes into automation:
   - add a doctor script (checks `package.json` exports/resolution, bun/npm mismatch, embedding provider validity)
   - add a “recommended transport” section: socket.io vs SSE with deployment requirements
5) **Token utility messaging**: consolidate into one canonical statement tied to concrete product surfaces (Jeju gas, product currency, buyback mechanism) and publish a roadmap of when each becomes observable by users.