autolab-hermes-delegation
Use Hermes delegate_task cleanly in this repo for planner, reviewer, researcher, reporter, experiment-worker, and memory-keeper roles.
69
56%
Does it follow best practices?
Impact
100%
5.00xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Fix and improve this skill with Tessl
tessl review fix ./projects/pre-training/.agents/skills/autolab-hermes-delegation/SKILL.mdEvaluation results
100%
91%Autolab Orchestration Script
Planner-reviewer delegation setup
Planner toolset file-only
0%
100%
Reviewer toolset file-only
0%
100%
Planner max_iterations
0%
100%
Reviewer max_iterations
0%
100%
Planner goal content
37%
100%
Planner context files
0%
100%
Planner single-change constraint
25%
100%
Reviewer goal content
25%
100%
Reviewer priority checks
25%
100%
Parent toolset includes delegation
0%
100%
No Hermes memory in toolsets
0%
100%
AGENTS.md only rulebook
0%
100%
100%
70%Research and Fleet Monitoring Script
Researcher-reporter delegation config
Researcher toolset complete
100%
100%
Researcher max_iterations=30
0%
100%
Researcher goal paper-derived
75%
100%
Researcher context files
0%
100%
Researcher rejects known ideas
50%
100%
Reporter toolset correct
50%
100%
Reporter max_iterations=25
0%
100%
Reporter credentials command
0%
100%
Reporter trackio summary command
0%
100%
Reporter no-edit constraint
50%
100%
Reporter source of truth
0%
100%
100%
79%Experiment Worker Orchestration
Experiment worker and memory-keeper flow
Worker create command
0%
100%
Worker delegate command
0%
100%
Experiment-worker toolset
0%
100%
Worker env var exports
0%
100%
Worker worktree cd
12%
100%
Train.py only edit
90%
100%
Worker submit_patch.py
0%
100%
Memory-keeper toolset file-only
12%
100%
Memory-keeper main checkout only
100%
100%
Memory-keeper preserved fields
20%
100%
Memory-keeper max_iterations=25
0%
100%
- Repository
- huggingface/context-course
- Commit
0448a7c
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
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.