Heroes of AI League (HAIL)

By Razvan Bradea

Overview

HAIL interviews an engineering team on AI adoption and system health, then places the team on a hero archetype. In ChatGPT, an 8-step wizard shows one question at a time with three answer panels. The team name is inferred. The same interview works in text only if the wizard does not render. The result includes the archetype, two axes, next actions, and a share link. The detailed What to Fix playbook is not available in this chat.

Tools

get_playbook

ChatGPT

Get the What-to-Fix playbook for a run the caller already owns. Requires a Save token proving ownership — run_id alone is not enough. When the playbook is locked, the response says so. This tool only reads. Do not print a locked playbook and do not include a checkout URL.

get_result

ChatGPT

Fetch a stored HAIL result by run_id or share_slug.

list_questions

ChatGPT

List HAIL assessment questions for simple, professional, or advanced profile. Advanced (real-number telemetry) is always available. Each question includes tier (core|extended) and core_position. Payload also returns core_count, core_keys, and extended_order.

run_assessment

ChatGPT

Score a team assessment and store a new run. Email is optional — anonymous runs are allowed. profile may be simple, professional, or advanced (advanced is always available; answers for advanced are numeric measurements, not option indices). Core 8 answers are enough to place a hero; response includes core_complete and sharpen hints. Agents are bounded by the per-IP rate limit, not a free-run gate. team_name is stored as Team DDMMYYYY- {intent}. The reply must include share_url in the same message as the archetype. This tool does not sell anything. Do not mention checkout or an upgrade URL.

show_assessment

ChatGPT

Open an 8-step wizard for the simple core assessment. One question per step: the Business-test question, its explanation, three answer panels, and I don't know. Use that text, explanation, and those labels exactly. Do not reword them. I don't know is the last choice and is left out of the score. The strongest answer is the third panel. Pass team_name chosen from the conversation; do not ask the user for one and do not collect it in the widget. Does not score or store a run. The wizard calls run_assessment after the eighth answer. After this tool returns, do not list the questions or options in the chat.

Capabilities

WritesInteractive

Example Prompts

Click any prompt to copy it.

Links

Documentation Privacy Policy

Similar Apps

AI Roleplay Chat Simulator

AI Detector by AI GPT Zero