Echo RunPod
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| RUNPOD_API_KEY | Yes | Your RunPod API key. Secret: vault://runpod/api-key then RUNPOD_API_KEY. |
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": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| searchA | Search local files/directories and/or Echo SDK capability registry (mcp__claude_ai_EchoOmegaPrime__search). target=files|sdk_caps|both. For red/unregistered grounding: target=sdk_caps, query='red unregistered', optional health=red|unknown|unregistered, check_ids for exact registration probes. |
| readC | Read a local file/directory or read one exact SDK capability record (includes registration_state). |
| writeA | Write or append UTF-8 text to a local file path. Idempotent via idempotency_key (or content-hash for overwrite). No delete operation is exposed. |
| invoke_sdk_capA | Invoke an Echo SDK capability by exact capability ID with params. Risky caps require confirm='SDK_INVOKE_APPROVED' unless ECHO_MCP_ALLOW_RISKY_SDK=1. Supports idempotency_key + gate retry/backoff. |
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 4 tools
search (query-based discovery) vs read (fetch an exact path/ID) vs write vs invoke_sdk_cap are largely distinct actions, and descriptions clarify the file-vs-SDK-capability duality. However, search and read both span files and SDK capabilities, creating mild overlap an agent could occasionally confuse when deciding whether to look up or fetch a record.
Three tools use terse single verbs (search, read, write) while one uses a fuller verb_noun form (invoke_sdk_cap); all are lowercase snake_case and verb-first. The pattern is readable and mostly consistent, with the one longer name being the only deviation.
Four tools is compact but each earns its place: discovery, read, write, and remote capability invocation cover the core surface of a file+SDK server. It is on the thin side but defensible for the focused scope.
The set exposes read, write/append (idempotent overwrite), search, and SDK invocation, but no delete or directory/list mutation is offered despite the file-oriented surface. Missing delete and any explicit listing/management operation is a notable gap agents will hit when housekeeping files.