Keep what each engine can do in one table instead of Claude checks spread through the code
Maintainers usually reply within 2 days
Assessment
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Newbie friendliness
- 48/100
- Issue type
- Feature
- Clarity
- Mostly clear
- Activity status
- Active
- Tech stack
- typescript
- Domain
- ai-infra-agents, testing, tooling
Research direction
Start by reading the engine presets and the existing capability checks named in the issue, especially isClaudeProvider and the fields hiveAware, canReceiveInbox, resumeFlag, and supportsModel. Then locate BUILD_ENGINES and the existing tests around engine configuration. Done means a typed capability table covers each engine, engineGaps(engine, os, needs) reports missing capabilities, and a test fails when an engine lacks an entry.
Written by the indexing model from the issue text.
Description
Problem / motivation
I was reading how workers get spawned and noticed that what Claude can do and other engines can't is decided inline, in many places.
isClaudeProvider(...)has 17 call sites in 8 files (hive.ts, index.ts, AddAgentModal, EditAgentModal, OnboardingWizard, ResearchPrivacySettings, useHive, store/config.ts).index.tsandhive.tseach put the result in a localclaudeProviderand branch on it again and again. In the spawn code inindex.tsit gates connectors, the per-agent scratch folder, the Claude-only flags and resume, the permission mode, the mailbox MCP, recording the Claude Code version, and Research refusing a non-Claude engine.- Four more places compare
provider === 'claude'directly (CommandCenterPanel, store/config.ts, providerAutomation.ts, teamAdvice.ts). - A few things are proper preset fields (
hiveAware,canReceiveInbox,resumeFlag,supportsModel), and some gaps only exist as open issues (#298, #414, #357, #109).
So if someone adds Codex to BUILD_ENGINES (still ['claude']), they have to find every one of these by hand and work out what that engine is missing. #48 and #139 cover this for folder privacy. The same problem exists for everything else.
Closest are #48, #139 (closed as a dupe of #48), #443 and #83. They cover enforcement and testing. This is the general table.
Proposed solution
A small table, one entry per engine, saying what it supports: system prompt on resume, hooks, tool guards, file sandbox, context gauge, resume, which OSes, which versions we've tested. Filled in from the preset fields we already have, and unknown where nobody has checked.
Plus a pure engineGaps(engine, os, needs) that returns what an engine is missing, and a test that fails if an engine is added without an entry.
I'd start with just the types, the table, the function and the test. No behavior change and no call sites touched. Replacing the checks can come later, one PR at a time. This could also be where #48's enforces record lives.
Screenshot, mockup or recording
No response
Alternatives considered
Leave it as is, or do it as part of #48. I'd rather not tie this small step to the bigger compiler change.
Area
The hive (memory, routing, GOD agent)
Pre-flight
- I attached a screenshot, mockup or recording above.
- I searched existing issues and discussions, and this isn't a duplicate.
- Dominant language
- TypeScript
- Stars
- 273
- Forks
- 211
- Avg merge
- 2d 9h
- Merged PRs (30d)
- 367
Getting set up
- No Dockerfile or Docker Compose file
- Has a pull request template
- Read the contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from agentvivekkumar/dontbemichael
-
auto-report
Difficulty 4/5 3-5 days Newbie friendliness 35/100
agentvivekkumar/dontbemichael#667 · 2 comments ·
Maintainers usually reply within 2 days
-
auto-report
Difficulty 4/5 3-5 days Newbie friendliness 35/100
agentvivekkumar/dontbemichael#665 · 2 comments ·
Maintainers usually reply within 2 days
-
auto-report
Difficulty 4/5 1-2 days Newbie friendliness 35/100
agentvivekkumar/dontbemichael#661 · 3 comments ·
Maintainers usually reply within 2 days
-
Difficulty 5/5 Over a week Newbie friendliness 28/100
agentvivekkumar/dontbemichael#607 · 1 comment ·
Maintainers usually reply within 2 days
-
Give Fable a tier and a pricePossibly taken @sahibsiddiqui claimed this 11 days ago. Openenhancement
Difficulty 2/5 1-3 hours Newbie friendliness 15/100
agentvivekkumar/dontbemichael#556 ·
Maintainers usually reply within 2 days
All issues in agentvivekkumar/dontbemichael
Similar issues
-
docker infra P1 security size:S
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
josephfung/curia#2408 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 62/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 65/100
ehmpathy/rhachet-roles-ehmpathy#1231 ·
Maintainers usually reply within 1 day
-
enhancement good first issue help wanted
Difficulty 2/5 Half a day Newbie friendliness 70/100
danielw-sudo/agentuto#32 ·
Maintainers usually reply within 1 day
-
type:docs
Difficulty 2/5 1-3 hours Newbie friendliness 65/100
nightgauge/nightgauge#2836 ·
Maintainers usually reply within 1 day