trajrl 0.3.2

Official skill collection for TrajectoryRL (Bittensor SN11) — discover and run subnet skills.

trajrl

The official CLI for TrajectoryRL — an open skill factory that leverages Bittensor's distributed compute and incentive layer with reinforcement learning to produce state-of-the-art agent skills.

One install gives any human or AI agent (Claude Code, Cursor, Codex, OpenClaw, Hermes, Manus, …) access to every skill TrajectoryRL has shipped. Each skill is a self-contained SKILL.md that agents can discover and follow directly.

CLI output is JSON when piped, Rich tables when interactive.

Install

pip install trajrl

Skills

Skills are the core of trajrl. Each one is a self-contained SKILL.md providing everything an agent needs: context, CLI commands, and data concepts. The skill catalog is fetched on demand so users always get the latest set as the subnet ships new winners.

subnet-analyze

Deep analysis of live TrajectoryRL data — validators, submissions, scores, weight distribution, scenarios, and eval logs.

What an agent can do with this skill:

# Full validator analysis — scores, weights, scenarios, leaderboard
trajrl subnet analyze 5FFApaS7...

# Drill into top submissions
trajrl subnet analyze 5FFApaS7... --deep

# Network overview
trajrl subnet status

# JSON output for piping (automatic when piped)
trajrl subnet status | jq '.validators.validators[].hotkey'

See skills/subnet-analyze/SKILL.md for full usage reference.

Links

• Subnet repo: https://github.com/trajectoryRL/trajectoryRL — incentive mechanism, evaluation framework, reference implementation
• Website: https://trajrl.com — leaderboard, live subnet data, dashboards
• Public API spec: PUBLIC_API.md — read-only, no auth, base URL https://trajrl.com
• Incentive mechanism: INCENTIVE_MECHANISM.md
• Evaluation dataset: DATASET_v0.1.md