dsh-mediacrawler
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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 |
|---|---|
| checkB | Check MediaCrawler source, runner, and optionally its full CLI dependencies. |
| collectB | Start one bounded collection run and return its durable run_id. |
| statusA | Get lifecycle state, outcome, and current artifact counts for a run. |
| runsA | List recent durable runs so an agent can recover their run IDs. |
| resultC | Get run status, typed artifacts, and a bounded redacted result sample. |
| delete_runA | Permanently delete one completed run after explicit confirmation. |
| cleanupC | Preview or delete completed runs outside a bounded retention window. |
| stopB | Idempotently stop a running crawler process tree. |
| logsA | Read redacted run logs using an exclusive sequence cursor. |
| artifactsC | List JSONL artifacts produced by a run using opaque artifact IDs. |
| previewA | Preview credential-redacted JSONL without accepting arbitrary file paths. |
| exportB | Create a credential-redacted, non-anonymized ZIP of run artifacts. |
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 12 tools
Most tools map to distinct lifecycle actions: check verifies the environment, collect launches a run, status/result monitor runs, and logs/artifacts/export/preview handle outputs. The only mild overlap is status vs result and preview vs result, but the descriptions clarify their different use cases.
The tool names mix imperative verbs (check, collect, stop, export) with noun endpoints (status, runs, result, logs, artifacts), and delete_run uses a verb_noun pattern that others do not follow. This is readable but not a predictable verb_noun convention across the set.
Twelve tools is well-scoped for a crawler run lifecycle, covering environment checking, execution, monitoring, artifact access, and cleanup without bloat. Each tool contributes a distinct stage, so the count feels appropriate.
The surface covers the full run lifecycle: preflight check, start, monitor, list, retrieve results, read logs/artifacts, export, stop, cleanup, and delete. No obvious critical operations are missing, and even retention cleanup and credential redaction are addressed.