draft-assistant
Skill91 starsChanged 2 months agoMITView on GitHub
Compare legal draft candidates and run continuous simulations with the manager MCP. Route live ESPN draft requests to the included ESPN companion workflow.
In this repository4 other files it gives its agents.
- AGENTS.mdAGENTS.md
- espn-automationSkill
- portfolio-managerSkill
- season-managerSkill
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 2 are distinctive, and 8 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 10
Read get_capabilities and get_manager_config before selecting the workflow.
- Very specific: line 17
Use season-manager after an explicit season connection.
- Very specific: line 23
Use load_demo(mode="draft") for a requested synthetic demonstration.
- Very specific: line 24
Use import_league_snapshot for supplied JSON observations.
- Very specific: line 26
Call recommend_draft for a bounded calculation.
- Very specific: line 27
Use start_draft_monitor for continued calculations from imported observations.
- Very specific: line 28
Read get_draft_recommendations for current results.
- Very specific: line 29
Use stop_draft_monitor when that monitoring task ends.
- Distinctive: line 33
Use a separate data directory for demonstrations when a real league is loaded.
- Distinctive: line 45
Do not substitute synthetic execution for a requested live draft action.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Draft assistant
The plugin includes the manager MCP and ESPN companion MCP.
Read get_capabilities and get_manager_config before selecting the workflow.
For a real ESPN draft, use the included espn-automation skill and its companion tools.
The manager's execute_demo_action tool is for synthetic state only.
The published v0.4.0 preview preserves the browser draft adapter.
The source browser extra and installed Chrome are required for that adapter.
Its new HTTP transport applies to season operation and does not replace draft submission.
Use season-manager after an explicit season connection. Completing a draft does not authorize new season action modes.
Distinguish historical live draft evidence from new source changes and fictional tests.
Imported drafts and demonstrations
- Identify the league, team, season, and current board revision.
- Use
load_demo(mode="draft")for a requested synthetic demonstration. - Use
import_league_snapshotfor supplied JSON observations. - Preserve the expected revision when replacing existing state.
- Call
recommend_draftfor a bounded calculation. - Use
start_draft_monitorfor continued calculations from imported observations. - Read
get_draft_recommendationsfor current results. - Use
stop_draft_monitorwhen that monitoring task ends.
The imported-state monitor does not read a browser. Use espn_start_automation for live ESPN observation.
load_demo refuses to replace imported real state or reset confirmed picks.
Use a separate data directory for demonstrations when a real league is loaded.
Report estimates
Report source age, board revision, actual trial count, and legal alternatives. Explain that availability is conditional on the modeled opponent selections. The score compares the next two selections. It is not a championship probability. More trials do not repair an incomplete board or incorrect projection. Strategy preferences cannot override roster rules or action limits.
For synthetic actions, use prepare_action and execute_demo_action within the returned policy requirements.
For live ESPN picks, use espn_prepare_draft_pick, espn_submit_draft_pick, and espn_reconcile_draft_pick.
Do not substitute synthetic execution for a requested live draft action.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

