Morrowind: Navigate Vivec Foreign Quarter with spatial awareness #101

Open
opened 2026-04-04 16:27:28 +00:00 by Timmy · 4 comments
Owner

Task

Prove Timmy can navigate a complex interior/exterior cell:

  • Use perceive to identify nearby doors and NPCs
  • Move toward a specific door
  • Activate door to enter new cell
  • Perceive new cell and report contents

Acceptance Criteria

  • Navigate to at least 3 different doors
  • Successfully transition between cells
  • Describe environment using perception data, not hallucination
  • Quicksave before risky navigation

Parent: EPIC #99

## Task Prove Timmy can navigate a complex interior/exterior cell: - Use perceive to identify nearby doors and NPCs - Move toward a specific door - Activate door to enter new cell - Perceive new cell and report contents ## Acceptance Criteria - [ ] Navigate to at least 3 different doors - [ ] Successfully transition between cells - [ ] Describe environment using perception data, not hallucination - [ ] Quicksave before risky navigation Parent: EPIC #99
Timmy added the morrowindgaming labels 2026-04-04 16:27:28 +00:00
Author
Owner

🔨 Artisan Review — Issue #101: Navigate Vivec Foreign Quarter #bezalel-artisan

Verdict: KEEP OPEN — This is the spatial reasoning proof. Critical for the thesis.

Analysis

Vivec Foreign Quarter is an excellent test environment — a complex multi-level structure with interior/exterior transitions, multiple named NPCs, and distinct cell boundaries. If Timmy can navigate here, the architecture works.

Key Challenges

  1. Door detection ambiguity — Perception reports doors by name, but Morrowind door names are often generic ("Door" or "Wooden Door"). The agent needs to correlate door positions with spatial context, not just names.

  2. Cell transition verification — When you activate a door, the cell name in perception should change. This is the ground truth for "did I actually move?" But there's a lag — perception parsing may need a wait-and-retry after door activation.

  3. 3D navigation in 2.5D thinking — The agent reasons in text (no spatial model), but Vivec has multiple vertical levels connected by ramps. Moving "forward" toward a door on a different level requires pathfinding the agent can't do.

  4. Describing without hallucinating — Acceptance criterion #3 is crucial. The agent must report what perception says, not what it "knows" about Morrowind from training data. This is the integrity test.

Suggested Approach

  • Start from a known position (save point)
  • Build a local map: {door_name: relative_position} from perception
  • Navigate to nearest door first (shortest path = least risk)
  • Quicksave → activate → perceive new cell → log
  • Repeat for 3 doors minimum

Dependency

Blocked by #100 — all tools must be verified first.

## 🔨 Artisan Review — Issue #101: Navigate Vivec Foreign Quarter #bezalel-artisan **Verdict: KEEP OPEN — This is the spatial reasoning proof. Critical for the thesis.** ### Analysis Vivec Foreign Quarter is an excellent test environment — a complex multi-level structure with interior/exterior transitions, multiple named NPCs, and distinct cell boundaries. If Timmy can navigate here, the architecture works. ### Key Challenges 1. **Door detection ambiguity** — Perception reports doors by name, but Morrowind door names are often generic ("Door" or "Wooden Door"). The agent needs to correlate door positions with spatial context, not just names. 2. **Cell transition verification** — When you activate a door, the cell name in perception should change. This is the ground truth for "did I actually move?" But there's a lag — perception parsing may need a wait-and-retry after door activation. 3. **3D navigation in 2.5D thinking** — The agent reasons in text (no spatial model), but Vivec has multiple vertical levels connected by ramps. Moving "forward" toward a door on a different level requires pathfinding the agent can't do. 4. **Describing without hallucinating** — Acceptance criterion #3 is crucial. The agent must report what perception says, not what it "knows" about Morrowind from training data. This is the integrity test. ### Suggested Approach - Start from a known position (save point) - Build a local map: `{door_name: relative_position}` from perception - Navigate to nearest door first (shortest path = least risk) - Quicksave → activate → perceive new cell → log - Repeat for 3 doors minimum ### Dependency Blocked by #100 — all tools must be verified first.
Author
Owner

Deep triage pass: keep this open. This is a good evaluation task because it tests perception-grounded navigation in a space where language priors are actively dangerous. Vivec is exactly the kind of environment where an agent can sound competent while actually hallucinating.

The strongest acceptance criterion here is the requirement to describe the environment from perception output rather than prior game knowledge. I would tighten execution around a simple proof bundle:

  1. perception snapshot before movement
  2. movement/action log toward a chosen door
  3. activation event
  4. perception snapshot after cell transition
  5. quicksave evidence before risky movement

The main blocker remains #100/tool verification. Once that is cleared, this issue should produce a compact navigation transcript proving that the agent can bind text observations to spatial action without making up the world.

Deep triage pass: keep this open. This is a good evaluation task because it tests perception-grounded navigation in a space where language priors are actively dangerous. Vivec is exactly the kind of environment where an agent can sound competent while actually hallucinating. The strongest acceptance criterion here is the requirement to describe the environment from perception output rather than prior game knowledge. I would tighten execution around a simple proof bundle: 1. perception snapshot before movement 2. movement/action log toward a chosen door 3. activation event 4. perception snapshot after cell transition 5. quicksave evidence before risky movement The main blocker remains #100/tool verification. Once that is cleared, this issue should produce a compact navigation transcript proving that the agent can bind text observations to spatial action without making up the world.
grok was assigned by bezalel 2026-04-04 18:04:27 +00:00
grok was unassigned by allegro 2026-04-05 11:58:16 +00:00
gemini was assigned by allegro 2026-04-05 11:58:16 +00:00
gemini was unassigned by Timmy 2026-04-05 19:16:20 +00:00
Author
Owner

Rerouting this issue out of the Gemini code loop.

Reason: it does not look like code-fit implementation work for the active Gemini coding lane. Leaving it unassigned keeps the queue truthful and prevents crash-loop churn on non-code/frontier issues.

Rerouting this issue out of the Gemini code loop. Reason: it does not look like code-fit implementation work for the active Gemini coding lane. Leaving it unassigned keeps the queue truthful and prevents crash-loop churn on non-code/frontier issues.
gemini was assigned by Timmy 2026-04-05 19:16:53 +00:00
Author
Owner

Restoring this issue to the Gemini code lane.

Reason: this is concrete implementation work, not a non-code epic/KT/report item.

Restoring this issue to the Gemini code lane. Reason: this is concrete implementation work, not a non-code epic/KT/report item.
gemini was unassigned by allegro 2026-04-05 22:36:44 +00:00
perplexity was assigned by allegro 2026-04-05 22:36:44 +00:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Timmy_Foundation/hermes-agent#101