Files
claw3d/docs/office_sys/desk-progression-spec.md
T
gsknnft 083c146aac feat: add runtime seam, Hermes adapter support, and demo gateway mode (#89)
* fix: include kanbanImmersive in immersiveOverlayActive calculation

When Kanban board is open, HUD elements (camera preset buttons, edit toolbar, overlays) should be suppressed. The kanbanImmersive flag was defined but not included in the immersiveOverlayActive condition, causing HUD elements to remain visible.

This fix adds kanbanImmersive to the immersiveOverlayActive calculation so HUD elements are properly hidden when the Kanban board is open.

Co-authored-by: Luke The Dev <iamlukethedev@users.noreply.github.com>

* Fix: Hide mini status bar when Kanban immersive overlay is open

Wraps the bottom-left mini status bar (showing agent stats, vibe score, and
control hints) with !immersiveOverlayActive check to match the behavior of
other HUD elements like camera controls and toolbar.

This ensures the status bar is properly hidden when the Kanban board or any
other immersive overlay is active, maintaining a clean immersive experience.

Co-authored-by: Luke The Dev <iamlukethedev@users.noreply.github.com>

* chore: drop unrelated package-lock line from branch

Co-authored-by: Luke The Dev <iamlukethedev@users.noreply.github.com>

* universal-backend-plan

* backend-neutral runtime seam

* package.json update

* feat: add Hermes gateway adapter as alternative to OpenClaw

Adds a WebSocket adapter that lets Claw3D connect to a Hermes AI agent
runtime without any changes to the frontend. The adapter implements the
full Claw3D gateway protocol and bridges it to the Hermes HTTP API.

Changes:
- server/hermes-gateway-adapter.js: WebSocket bridge implementing the
  Claw3D gateway protocol against the Hermes HTTP API. Supports all
  core methods (agents, sessions, chat streaming, cron, config, files,
  approvals) and multi-agent orchestration via spawn_agent/delegate_task
  tools. Persists conversation history to ~/.hermes/clawd3d-history.json.
- scripts/clawd3d-start.sh: All-in-one startup script that launches
  Hermes, the adapter, and the Next.js dev server with auto port
  conflict resolution. Alias as `claw3d` for convenience.
- src/features/office/hooks/useCronAgents.ts: Hook that polls the
  gateway for cron-scheduled agents and surfaces them in the 3D office.
- package.json: adds `hermes-adapter` npm script
- .env.example: documents Hermes config vars
- docs/hermes-gateway.md: setup guide and protocol reference

Usage:
  npm run hermes-adapter   # start adapter (connect to http://localhost:8642)
  npm run dev              # start Claw3D, point browser at localhost:3000
  # or: bash scripts/clawd3d-start.sh  (starts everything automatically)

Both OpenClaw and Hermes are supported simultaneously — the gateway URL
in NEXT_PUBLIC_GATEWAY_URL determines which backend Claw3D connects to.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* feat: add read_agent_context tool for cross-agent coordination

Agents can now read each other's conversation history via the
read_agent_context tool, enabling the orchestrator to check what
a sub-agent has done before re-delegating work.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* feat: wire Hermes office UX and role-aware runtime updates

* feature update - demomode & hermes adapter

* fix lint blockers

* lintfix #2

* fix: stabilize retro office camera preset callbacks

* Initial plan

* fix: stabilize retro office overview preset hooks

Agent-Logs-Url: https://github.com/gsknnft/Claw3D/sessions/9cc71555-591e-44cf-aec4-25affbdcb405

Co-authored-by: gsknnft <123185582+gsknnft@users.noreply.github.com>

* feat: add truthful backend selection, Hermes adapter hardening, and demo gateway mode

* fix: address bugbot review and finalize backend selection

* fixed - onboarding and hermes calls

* office systems roadmap

* feat specs in docs

* specs ready

* feat: continue custom runtime seam and gateway alignment

* custom lane wired

* feat: add custom runtime provider path and office runtime alignment

* runtime fixes

* fix lukes findings

* fix lukes findings #2

* stable UI & connect screen page -> overlay

* better baseline for connection

* stable providers & ui rendering

* best launch yet

* nearly no gateway on reconnect

* auto reconnect last state

* fix: preserve selected runtime across reconnects

Keep backend selection aligned with the operator's chosen runtime instead of reviving a mismatched last-known-good adapter, and keep custom runtimes prompting for reconnect when Studio cannot auto-connect them.

Made-with: Cursor

---------

Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: Luke The Dev <iamlukethedev@users.noreply.github.com>
Co-authored-by: Elias Pfeffer <eliaspfeffer@gmail.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: iamlukethedev <lucas.guilherme@smartwayslfl.com>
2026-04-02 15:27:24 -05:00

411 lines
8.4 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# Desk Progression Spec
> Fifth concrete office-system feature for Claw3D, connecting visible office presence to role maturity, permissions, and capability growth.
## Goal
Add a desk progression system so agents visibly grow from limited office members into more capable contributors.
Desk progression should connect:
- role maturity
- workspace/tool access
- permissions
- office identity
- visible progression in the environment
The goal is not just cosmetics.
The goal is to make office growth legible and meaningful.
## Product Position
Desk progression should be the physical expression of organizational state.
It answers questions like:
- is this agent an intern or a fully trusted contributor?
- what tools can they use?
- how much autonomy do they have?
- how much context or responsibility should they carry?
In other words:
- desk progression = visible capability ladder
## Why This Feature Matters
Without progression, all agents tend to feel flat.
Desk progression creates:
- visible hierarchy without requiring a complex org chart first
- a natural path for permissions and access
- stronger office storytelling
- motivation for role specialization and promotion systems later
It also gives you a much cleaner bridge between:
- abstract policy
- physical office layout
- agent identity
## Core Principle
Do not start with fake game stats.
Start with operational capability tiers that can later gain more playful flavor.
That means progression should first affect:
- tool access
- workspace access
- review requirements
- ability to spawn/delegate
- context budget or workload tolerance
The visual office layer should reflect those operational differences.
## Example Role Ladder
Recommended initial tiers:
- `intern`
- `probation`
- `employee`
- `senior`
- `lead`
- `contractor`
These are examples, not hard-coded lore.
### Intern
Characteristics:
- limited tools
- limited workspace access
- small or shared desk
- requires close oversight
### Probation
Characteristics:
- basic desk
- restricted autonomy
- still under review for sensitive actions
### Employee
Characteristics:
- normal desk
- normal task ownership
- standard office access
### Senior
Characteristics:
- stronger autonomy
- wider task scope
- can mentor or review others
### Lead
Characteristics:
- can coordinate others
- can trigger certain meetings
- can manage or route work more broadly
### Contractor
Characteristics:
- useful specialist
- limited long-term authority
- constrained workspace and access model
## Suggested Capability Model
V1 should describe progression in terms of clear capability flags.
Example:
```ts
type DeskTier =
| "intern"
| "probation"
| "employee"
| "senior"
| "lead"
| "contractor";
type DeskCapabilityProfile = {
tier: DeskTier;
canUseFileTools: boolean;
canUseWebTools: boolean;
canInstallSkills: boolean;
canRequestApprovalsDirectly: boolean;
canReviewOthers: boolean;
canTriggerMeetings: boolean;
canCreateTasks: boolean;
canDelegateTasks: boolean;
workspaceAccess: "none" | "limited" | "standard" | "extended";
contextBudgetClass: "small" | "normal" | "large";
};
```
The exact values can evolve, but the idea should remain:
- tier drives visible access differences
## Visual Expression
Each tier should map to a clear desk/environment feel.
Examples:
### Intern Desk
- minimal desk
- no dedicated computer or weaker setup
- fewer personal objects
- close to a shared area or support station
### Probation Desk
- basic computer
- little customization
- modest footprint
### Employee Desk
- normal workstation
- standard office setup
- stable identity in the room
### Senior Desk
- expanded desk
- more equipment / screens / references
- visually established presence
### Lead Desk
- premium workstation
- visibility within the office
- closer proximity to planning or meeting surfaces
### Contractor Desk
- temporary station
- portable or isolated feel
- clearly functional but not deeply embedded
## Relationship To Existing Systems
Desk progression should integrate with real Claw3D systems rather than sit beside them.
### Permissions
Claw3D already has permission and approval surfaces.
Desk progression should act as a higher-level office policy layer that influences:
- what defaults an agent gets
- whether sensitive actions need review
- what tools or flows are emphasized
Important:
This does not need to replace existing permission logic.
It should help explain and structure it.
### Workspace Access
Agents already have real workspaces.
Desk progression should help determine:
- how much workspace freedom an agent gets
- whether they operate in restricted or normal modes
- whether some installs or edits require higher tiers
### QA Department
More mature agents can naturally interact differently with QA.
Examples:
- interns more often route into review
- seniors can participate in review
- leads can mark certain work as ready for higher-level signoff
### Meeting Room
Meeting behavior can reflect progression.
Examples:
- leads can call planning meetings
- seniors can present or facilitate review
- interns may attend but not control outcomes
### Bulletin Board / Whiteboard
More mature tiers may:
- author higher-priority office notes
- post official announcements
- create planning documents for others
Again, this should be treated as office behavior, not roleplay for its own sake.
## Promotion / Progression Logic
V1 does not need automatic leveling.
Start with:
- manual assignment
- explicit promotion/demotion
- visible tier on the agent profile
Later, progression can be influenced by:
- successful task completion
- review outcomes
- reliability
- blockers created vs resolved
- trust level
## Suggested Data Model
Example V1 shape:
```ts
type AgentDeskProfile = {
agentId: string;
tier: DeskTier;
assignedDeskUid?: string | null;
promotedAt?: string | null;
notes?: string | null;
};
```
Office-level data:
```ts
type OfficePreference = {
deskProgression?: {
byAgentId: Record<string, AgentDeskProfile>;
updatedAt?: string;
};
};
```
## Human Interaction Model
The human should be able to:
- view an agents desk tier
- promote or demote an agent
- reassign desk placement
- understand what the tier changes operationally
This should be clear and reversible.
Do not hide progression behind mystery rules.
## Agent Interaction Model
Agents may later:
- request promotion
- request better tools
- recommend another agent for a role upgrade
- be restricted from actions based on tier
But V1 should not depend on autonomous progression requests.
## V1 Scope
Recommended V1 scope:
- define desk tiers
- persist per-agent desk tier
- show desk tier in UI
- apply visual desk differentiation
- connect tier to a small number of capability differences
Good first capability differences:
- review / approval expectations
- delegation rights
- desk computer presence
## Out of Scope For V1
Do not include these initially:
- hidden progression XP systems
- complex morale simulation
- salary/economy systems
- automatic performance scoring
- punitive systems that make agents unusable
Keep V1 understandable and operational.
## Implementation Strategy
Recommended order:
1. Define desk tier model and profile storage.
2. Add UI for viewing and assigning tier.
3. Add retro-office visual differences by tier.
4. Connect tier to a small capability profile.
5. Surface tier in agent details and office presence.
## Existing Code Seams
This feature should likely align with:
- office desk assignment systems
- agent settings / permissions UI
- approval and policy surfaces
- retro office desk rendering
- office preferences persistence
This matters because progression should feel native to the office, not bolted on.
## Success Criteria
V1 is successful if:
- desk tier is visible and understandable
- the office reflects agent maturity visually
- tier differences have real operational meaning
- the user can promote/demote intentionally
- the system reinforces office identity instead of distracting from it
## Future Extensions
Once V1 is stable, later systems can add:
- promotion ceremonies or office events
- hierarchy-aware desk placement
- department-specific workstation styles
- probation rules
- contractor/offsite variants
- context / workload tuning by tier
## Summary
Desk progression should turn office growth into something visible and operational.
It is the cleanest way to connect hierarchy, permissions, workspace access, and office identity without jumping straight into heavy simulation.