Execution: human-only

5 of 30 records. Results are T2 and above.

What matches

a-prompt-the-filter-refuses

Reword a request a safety filter declined, in a fixed order, and keep the pair — because the refusal is usually about how it was asked rather than what was asked.

Publisher
adapted tier U
Targets
copilot-studio
Highest risk
read-only
Execution
human-only
Profile
P1
Trust
T2 · 2026-10-11 · runbooks.directory

copilot-agent-answers-wrong

Find out why a Copilot Studio agent is answering wrongly — whether the content is missing, unreachable, or being ignored — before changing anything.

Publisher
adapted tier U
Targets
copilot-studio
Highest risk
reversible-write
Execution
human-only
Profile
P1
Trust
T2 · 2026-10-11 · runbooks.directory

copilot-agent-withdraw-after-publish

Take a published Copilot Studio agent back out of service, least drastic action first, when it is answering badly in front of users.

Publisher
adapted tier U
Targets
copilot-studio
Highest risk
irreversible
Execution
human-only
Profile
P1
Trust
T2 · 2026-10-11 · runbooks.directory

copilot-credit-spike-contain

Stop a Copilot Studio agent burning credits faster than it was budgeted for, before deciding whether the consumption is genuine.

Publisher
adapted tier U
Targets
copilot-studio, power-platform
Highest risk
destructive
Execution
human-only
Profile
P1
Trust
T2 · 2026-10-11 · runbooks.directory

ci-refuses-to-start-deploy-by-hand

The pipeline refuses a run before it starts, so the deploy has to happen from a workstation — told apart from a real failure first, and with what the manual path costs put back afterwards.

Publisher
std tier U
Targets
github-actions, terraform, aws
Highest risk
irreversible
Execution
human-only
Profile
P1
Trust
T2 · 2026-10-11 · runbooks.directory