codex-advisor
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GEMINI_API_KEY | No | Gemini API key for using Gemini provider. | |
| OPENAI_API_KEY | No | OpenAI API key for using OpenAI API provider. | |
| ANTHROPIC_API_KEY | No | Anthropic API key for using Claude provider. | |
| OPENROUTER_API_KEY | No | API key for custom OpenAI-compatible endpoints (e.g., OpenRouter). |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| consult_advisorA | Consult the configured advisor model for strategic guidance. Call this at decision points: before committing to a non-trivial plan,
when the same error persists after two fix attempts, or before declaring
a complex task complete. The current session transcript is attached
automatically; put the specific decision to evaluate in |
| advisor_configA | Get or change advisor settings. Use when the user asks. action='get' shows current settings. action='set' changes the advisor model; pass model as '/', e.g. 'codex/gpt-5.6-sol', 'openai/gpt-5.2', 'anthropic/claude-opus-4-8', 'gemini/gemini-2.5-pro'. Any model name is accepted (no allowlist). action='off' disables consultations. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 2 tools
The two tools are completely distinct: one performs a consultation, the other manages configuration. No overlap or ambiguity exists between them.
Tool names use different patterns: 'consult_advisor' follows verb_noun, while 'advisor_config' is noun_noun. The inconsistency is noticeable but both names are clear and self-explanatory.
With only 2 tools, the set feels thin but is appropriate for the narrow scope of advisor consultation and configuration. It covers the essential actions without unnecessary bloat.
The domain of consulting and configuring an advisor is well-covered. A minor gap is the lack of consultation history or logging, but for the stated purpose (strategic guidance and settings) the surface is essentially complete.