CtrlK
BlogDocsLog inGet started
Tessl Logo

agent-raft-manager

Agent skill for raft-manager - invoke with $agent-raft-manager

Install with Tessl CLI

npx tessl i github:ruvnet/claude-flow --skill agent-raft-manager
What are skills?

28

Does it follow best practices?

Validation for skill structure

SKILL.md
Review
Evals

Evaluation results

94%

7%

Distributed Cluster Leader Election Simulator

Leader election protocol

Criteria
Without context
With context

Randomized election timeout

100%

100%

Heartbeat messages

100%

100%

Candidate state

100%

100%

Vote collection

100%

100%

Split vote backoff

44%

66%

Leader failure detection

100%

100%

Term tracking

75%

100%

Without context: $0.2615 · 1m 20s · 14 turns · 19 in / 4,237 out tokens

With context: $0.4704 · 1m 55s · 24 turns · 29 in / 6,423 out tokens

100%

Distributed Transaction Log with Compaction

Log replication and compaction

Criteria
Without context
With context

AppendEntries protocol

100%

100%

Log consistency check

100%

100%

Commit index advancement

100%

100%

State machine application

100%

100%

Snapshot creation

100%

100%

Log truncation after snapshot

100%

100%

Snapshot-based node recovery

100%

100%

Without context: $0.6043 · 2m 31s · 24 turns · 33 in / 8,034 out tokens

With context: $0.9983 · 4m 39s · 30 turns · 197 in / 16,340 out tokens

76%

-17%

Elastic Cluster Membership Manager

Dynamic membership changes and fault tolerance

Criteria
Without context
With context

Two-phase membership change

100%

35%

Quorum based on new config

100%

93%

Both add and remove supported

100%

100%

Network partition modeled

100%

100%

Consistency during partition

100%

88%

Automatic node recovery

83%

75%

Safety during concurrent changes

50%

50%

Without context: $0.5220 · 2m 38s · 22 turns · 28 in / 8,872 out tokens

With context: $0.7372 · 3m 30s · 25 turns · 35 in / 12,547 out tokens

Evaluated
Agent
Claude Code

Table of Contents

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.