#27 Support OpenAI as a fleet member LLM backend

[Anonymous]

Originally created by: kumaakh

Description

Add support for OpenAI (GPT-4, o-series) as an alternative LLM backend for fleet members, alongside Claude.

Motivation

OpenAI models are widely available and some users may prefer them for specific tasks or already have API access. Supporting OpenAI broadens the fleet's reach and allows mixed-model teams. Users should be able to mix and match — e.g., some members running Claude, others running OpenAI — within the same fleet.

Scope

MCP Server (src/)

• Member registration should allow specifying the LLM provider/backend
• execute_prompt should route to the appropriate backend CLI/API (Codex CLI or OpenAI API)
• provision_auth should support OpenAI API key provisioning (OPENAI_API_KEY)
• update_claude tool needs generalization (update_agent_cli or similar)
• OS command builders (buildPromptCommand, claudeVersion, etc.) need OpenAI equivalents
• Error classification (prompt-errors.ts) needs OpenAI-specific patterns
• Agent type/interface needs an llmProvider field

PM Skill (skills/pm/)

• Model selection logic (haiku→sonnet→opus escalation) needs OpenAI equivalents (gpt-4o-mini→gpt-4o→o3 or similar)
• CLAUDE.md templates reference Claude-specific behavior
• Doer-reviewer loop assumes Claude CLI session semantics
• Troubleshooting guide assumes Claude error patterns

Documentation

• All tool docs reference Claude CLI exclusively
• User guide needs multi-provider setup instructions
• Architecture doc needs provider abstraction layer description

Non-goals (v1)

• OpenAI as the PM's own LLM (covered by separate provider-for-PM work)
• Automatic provider failover between OpenAI and Claude

[Anonymous]

Originally posted by: kumaakh

Covered by Codex CLI support (already shipped). Codex CLI is OpenAI's official agentic CLI tool — fleet registers these members with llmProvider: 'codex'. OpenAI backend support is available today via codex provider.

[Anonymous]

Ticket changed by: kumaakh

• status: open --> closed