agent-v3-memory-specialist
Agent skill for v3-memory-specialist - invoke with $agent-v3-memory-specialist
Install with Tessl CLI
npx tessl i github:ruvnet/claude-flow --skill agent-v3-memory-specialist31
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillValidation for skill structure
Evaluation results
73%
20%Unified Memory Search Module
HNSW indexing and query routing
HNSW dimensions
0%
100%
HNSW efConstruction
100%
100%
HNSW M parameter
100%
100%
HNSW maxElements
0%
100%
Semantic query routing
100%
100%
Structured query routing
50%
60%
Dual store write
100%
70%
IMemoryBackend interface
0%
0%
MemoryEntry embedding field
80%
100%
AgentDBAdapter usage
0%
0%
Without context: $0.5677 · 2m 58s · 21 turns · 27 in / 11,230 out tokens
With context: $2.0589 · 7m 24s · 68 turns · 509 in / 22,263 out tokens
96%
34%Agent Memory Migration
Legacy backend migration to AgentDB
SQLite field extraction
100%
100%
SQLite embedding generation
100%
100%
Markdown originalFile metadata
0%
100%
Markdown migrationDate metadata
0%
100%
Markdown type metadata
0%
100%
Markdown embedding generation
100%
100%
AgentDB as target
0%
100%
Phased migration plan
100%
100%
Backward compatibility
100%
100%
MemoryEntry structure
100%
50%
Order by created_at
100%
100%
Without context: $0.9842 · 4m 12s · 43 turns · 50 in / 13,487 out tokens
With context: $1.8546 · 7m 3s · 65 turns · 69 in / 19,540 out tokens
87%
56%Adaptive Learning Pattern Store
SONA pattern storage and benchmarking
sonaMode field present
0%
100%
Valid sonaMode values
0%
100%
reward metadata field
100%
100%
trajectory metadata field
0%
100%
adaptation_time metadata field
0%
100%
Semantic retrieval query type
0%
37%
Retrieval filter by type
0%
50%
Retrieval limit: 5
100%
100%
Benchmark 1000 queries
0%
100%
Benchmark queriesPerSecond
100%
100%
Benchmark avgLatency
50%
70%
Without context: $0.8755 · 3m 55s · 36 turns · 43 in / 14,136 out tokens
With context: $1.0550 · 4m 41s · 34 turns · 38 in / 16,985 out tokens
- Evaluated
- Agent
- Claude Code
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.