Act as an AI copilot during an SBTM exploratory test session, automatically tracking time and logging observations.
Installation
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Act as an AI copilot during an SBTM exploratory test session, automatically tracking time and logging observations.
license
MIT
metadata
{"author":"AI","version":"1.0"}
SBTM Session Copilot Workflow
Follow these instructions when the user invokes the /sbtm-session workflow to execute a charter.
Initialize Session:
Ask the user which Charter they want to execute.
Read the selected Charter from test-management/sbtm/[coverage-area]/charters/.
Ask the user for the intended timebox duration (e.g., 60 minutes) and start the session.
The Execution Loop (Copilot Mode):
The user will periodically send you short messages about what they are doing (e.g., "setting up accounts," "exploring checkout," "found a crash on submit").
DO NOT interrupt their flow. Acknowledge their messages briefly, store the context, and optionally suggest a heuristic or edge case to try based on their current focus.
CRITICAL: Monitor the timestamps of the conversation implicitly.
Session Conclusion & T/B/S Automation:
When the user says they are done, or the timebox expires, you must generate the Session Report.
Read test-management/sbtm/SESSION_TEMPLATE.md for formatting.
Automatic Time Tracking: You must calculate the T/B/S metrics (Test Execution, Bug Investigation, Setup/Admin) based on the flow of the conversation. Estimate the percentages of the timebox they spent on setup vs exploring vs bug logging. Do NOT ask the user to provide these percentages; do it for them.
Compile the Coverage, Bugs & Issues, and Notes.
Save the Report:
Save the completed report to test-management/sbtm/[coverage-area]/sessions/SESSION-[YYYYMMDD]-[CHARTER-ID].md.
Instruct the user to run /sbtm-review with a Senior QA for debriefing.