Redis

Databases

Inspect keys, analyze memory, and manage your Redis instances through natural conversation — Neotask runs on OpenClaw.

What You Can Do

Inspect Keys and Values

Ask Neotask to find keys matching a pattern, check the value and TTL of a specific key, or count how many keys exist in a given keyspace. It runs the right Redis commands (SCAN, TTL, TYPE, GET) and returns results in a readable format without raw CLI output.

Analyze Memory Usage

Ask for a breakdown of memory usage by key prefix or data type. Neotask runs MEMORY USAGE samples, analyzes the keyspace with OBJECT ENCODING, and identifies which patterns are consuming the most space — useful for optimizing large caches.

Monitor Cache Performance

Get a real-time health snapshot: cache hit rate from keyspace stats, connected clients, commands per second, and whether any slow commands are being logged. Neotask interprets INFO ALL output and highlights what needs attention.

Manage Expiry and Eviction

Ask which keys have no TTL set (potential memory leaks), update the TTL on a set of keys matching a pattern, or review the current eviction policy and whether it matches your use case.

Debug Pub/Sub and Streams

Inspect active pub/sub channels and subscriber counts, list consumer groups on a Redis Stream, and check pending message counts — useful for debugging event-driven architectures that use Redis as a message broker.

Try Asking

  • "How much memory is the session: key prefix using?"
  • "Find all keys matching user:*:cart and show their TTLs"
  • "What is the current cache hit rate for this Redis instance?"
  • "Which keys have no expiry set?"
  • "Show me the top 10 largest keys by memory usage"
  • "What is the replication lag on my Redis replica?"
  • "List all consumer groups on the events stream and their pending counts"
  • "Set a 1-hour TTL on all keys matching temp:* that currently have no expiry"

    • Pro Tips

  • Always use SCAN-based patterns in your prompts rather than KEYS — Neotask will use SCAN by default so it does not block production Redis instances.
  • Ask for OBJECT ENCODING on large keys: a Hash stored as a ziplist is fine, but one stored as a hashtable may have unexpected memory overhead.
  • Pair memory analysis with your eviction policy: if you are using allkeys-lru, ask Neotask which keys have not been accessed recently and could be allowed to evict.
  • Use Neotask to set up a quick monitoring baseline: "tell me these five stats every 5 minutes" creates a simple observability loop without a full monitoring stack.
  • For Redis Cluster deployments, specify which shard you want to inspect or ask for a cluster-wide summary — Neotask handles slot-based routing transparently.

    • Works Well With