## Intel — 2026-01-03 (covering latest available signals through 2026-01-02)

### 1) Data Pattern Recognition

**Development velocity (GitHub: elizaos/eliza, Jan 2–3 window)**
- PRs: **1 opened / 1 merged** (PR #6301, license-year chore; **1 file changed**, **+1/-1**)
- Issues: **14 opened**, **0 closed**
- Contributors: **4 active**
- Issue author concentration: **borisudovicic** drove the visible product-facing issue burst (notably #6311–#6315)

**Trend signal**
- Output is currently **issue-heavy vs. code-heavy** (14 issues vs. 1 trivial PR). This typically indicates: (a) UX/product discovery phase accelerating, or (b) bottlenecked implementation capacity/review bandwidth.
- Core-dev bandwidth risk flagged: reviewer availability constrained (Stan reviewing PRs while ill; requests for reviews).

**Community engagement patterns (Discord, Jan 1–2)**
- High “vision & narrative” engagement: **agentic gaming (RuneScape→Zelda-like with agents)**, hyperscape quest/NPC tooling, “agentic economy” excitement, partners discussing infra cost-reduction agents.
- High “support load” engagement: repeated **token migration** questions (snapshot eligibility, Phantom/Tangem/WalletConnect, tokens not showing) with multiple **unanswered** FAQs.
- Dev curiosity but low closure: questions like “import entire codebase as agent?” and “use plugins not added to cloud?” mostly unanswered, indicating **docs/enablement gap**.

**Feature adoption / funnel proxy signals**
- Public agent sharing appears to be driving inbound traffic (“clicked a link on Twitter”), but the experience is mismatched to user state (issues #6312/#6313/#6314).
- Monetization/abuse-control pressure is rising: free credits reduction (#6315) + non-auth message limits (#6312) imply a need to protect costs and improve conversion.

**Pain point correlation across channels**
- **Onboarding friction**: Discord migration confusion + product issues proposing gating (2–3 free messages) suggest the top-of-funnel is active but leaky.
- **UX clarity**: “Public agent states” separation (#6313) aligns with community’s share-link discovery behavior.
- **Quality perception**: “chat summaries don’t make sense” (#6311) is a direct UX-quality complaint; poor summaries degrade trust and return usage, especially for shared/public agents.

---

### 2) User Experience Intelligence

**Feedback themes (categorized by impact)**

**A. Critical (blocks conversion / causes support churn)**
1. **Token migration reliability + clarity**
   - Symptoms: tokens not showing after transfer; uncertainty about post-snapshot purchases; wallet compatibility (Tangem); repeated “when deadline?” questions.
   - Current handling: redirected to channels / DMs; many questions unanswered → escalates public confusion.
   - Impact: reputational + support load + conversion risk.

2. **Public agent onboarding mismatch (unauthenticated vs. authenticated vs. owner)**
   - Proposed by #6313; supported by #6312 and discovery behavior (Twitter links).
   - Impact: first-session confusion, lower sign-up conversion, feature misuse (seeing toggles/edit when they shouldn’t).

**B. High (meaningfully degrades product usefulness)**
1. **Chat summaries quality (#6311)**
   - Users report summaries “don’t make much sense,” with a preference for more informative titles/recaps.
   - Likely affects: navigation, retrieval, and perceived intelligence of agents.

2. **Public agent social proof / discovery metrics (#6314)**
   - Request: show total chat/message count on public agent cards.
   - Impact: improves selection confidence, shareability, and marketplace dynamics.

**C. Medium (developer adoption / ecosystem growth)**
1. **Plugin capability clarity**
   - “Discord plugin compatibility?” “Can I use plugins not added to cloud?” “Which repo to contribute to?”
   - Impact: slows contributions and increases repetitive Q&A.

2. **Workflow robustness**
   - Core-dev: multi-step workflows with retry logic + parameter extraction; singleShot parameter support to avoid “action problems.”
   - Impact: improves agent reliability; reduces “it failed mysteriously” support.

**Sentiment**
- Mixed: strong excitement around agentic games/infra agents, but recurring frustration/confusion around migration and onboarding. Net sentiment is positive on vision, negative on immediate UX/support gaps.

---

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

#### Priority 0: Stabilize onboarding + reduce support fire
**Initiative:** “Public Agent Entry States + Soft Gate” (issues #6313 + #6312 + #6314 as one epic)
- **User impact:** Very high (improves first-run experience, increases conversion, reduces confusion)
- **Technical risk:** Medium (UI state handling + auth gating + analytics plumbing)
- **Dependencies:** auth state detection; routing; basic event tracking; design pass

**Recommended scope (2-week target)**
1. **State-specific UI**
   - Unauthenticated visitor: chat-only view; agent identity; share link; remove toggles/edit/fork.
   - Authenticated non-owner: Chat + Fork; creator attribution; no private/public toggle.
   - Owner: Chat + Edit + privacy toggle; future analytics placeholder.
2. **Soft gate after 2–3 messages** (unauth) with clear CTA to sign up.
3. **Add “chat count / usage” metric on public agent card** (lightweight social proof).
4. Instrument conversion funnel: `public_agent_view → first_message → gate_hit → signup_start → signup_complete → resumed_chat`.

**Resource allocation**
- 1 FE engineer + 0.5 design + 0.5 BE (auth/session/events) for ~1 sprint.

#### Priority 1: Fix “quality tells” that undermine trust
**Initiative:** “Chat Summary Quality Upgrade” (#6311)
- **User impact:** High (improves navigation and perceived intelligence)
- **Technical risk:** Medium (prompting + evaluation; possible model/provider variance)
- **Dependencies:** summary generation pipeline; UI display format

**Recommended approach**
- Replace generic summaries with a **structured title + 1–2 bullet recap** based on last N turns.
- Add a lightweight evaluation loop: sample 50 chats/day, measure “rename rate” and “reopen-to-find” behavior as proxies.
- Provide fallback: if confidence low, show “Chat with {AgentName} — {date}” rather than incoherent text.

**Resource allocation**
- 1 engineer (full-stack) for 3–5 days for MVP, then iterate.

#### Priority 2: Token migration communications as an operational product surface
**Initiative:** “Migration FAQ + Known Issues + Wallet Support Policy”
- **User impact:** Very high (reduces confusion, protects brand)
- **Technical risk:** Low–Medium (docs + minor site updates; possible wallet connect changes)
- **Dependencies:** definitive policy decisions (deadline, snapshot rules), support ownership

**Recommended actions**
- Publish a single canonical FAQ answering the currently repeated unanswered questions:
  - migration deadline (date/time/UTC)
  - post-snapshot purchase eligibility
  - Phantom vs MetaMask steps
  - Tangem support status; if unsupported, explicit workaround path
- Add migration-site banner linking to FAQ + “if tokens not showing” checklist.
- Triage whether **WalletConnect** is needed short-term vs. longer-term.

**Resource allocation**
- 0.25 PM/ops + 0.25 engineer (website/docs) + 1 support lead to keep it current.

#### Priority 3: Reliability work that reduces downstream support (core workflows)
**Initiative:** “Workflow Robustness: retry logic, parameter extraction, singleShot parameters”
- **User impact:** Medium–High (fewer failed actions, better tool execution)
- **Technical risk:** Medium–High (edge cases, backwards compatibility)
- **Dependencies:** PR review bandwidth; test harness coverage

**Recommended sequencing**
- Unblock PR review capacity first (assign secondary reviewers; set SLA for PR feedback).
- Implement **singleShot parameter support** (explicitly referenced as preventing “action problems”).
- Add minimal regression tests around action invocation and parameter passing.

**Resource allocation**
- 1 platform engineer + 1 reviewer; prioritize review throughput.

---

## Key Metrics to Track Next (to validate strategy)
1. **Public agent funnel**
   - View→message rate, message→signup rate, signup completion, post-signup continuation.
2. **Support load**
   - # migration questions/day in Discord; % answered; median time-to-answer.
3. **Chat summary usefulness**
   - Chat rename frequency; bounce rate in chat list; user-reported confusion incidents.
4. **Engineering throughput**
   - Issues opened vs. closed weekly; PR cycle time; reviewer latency.

---

## Immediate Recommendations (next 72 hours)
- Convert #6312/#6313/#6314 into a single **tracked epic** with acceptance criteria and owner.
- Publish a **migration “top 5 questions”** answer in one pinned Discord post + docs link.
- Assign explicit reviewer coverage for streaming/workflow PRs (reduce single-point review dependency).
- Decide policy on **free credits** (#6315) only after message-limit gating and funnel telemetry are in place (otherwise risk harming acquisition without knowing CAC/activation impact).