A reusable skill that forgets an investigation's scope is incomplete. Knowing how to run a procedure helps little when its next invocation inherits vague permission. I want to reuse the capability without reconstructing from memory the conditions that made using it acceptable.
On August 11, 2026, I requested delegated research that could become a skill. The idea was to take a capability out of improvisation and make it available again. On August 10, OpenAI had detailed its Daybreak expansion, separating Blue and Red access and introducing GPT-5.6-Cyber for approved defenders. The announcement put capability alongside access verification and containment. That combination deserves attention.
My research request was no Daybreak execution. It illustrates the organizational problem I care about: delegating an investigation while the main work continues. When a capability concerns security, its instructions need to specify where to look. Knowing a technique does not grant an agent permission to use it on every system where it could help.
I would begin such a skill with the authorized targets and expected result. The environment needs to prevent an attempt from going beyond those targets. Copying a procedure from an earlier assignment requires reviewing that part, because the old authorization may have ended with the work. The procedure can remain useful while the permitted destination changes completely.
In July, I had already asked for autonomy within limits. For me, that includes distinguishing investigation from applying a change. Both stages can be authorized at the start. Where they are not, the research ends in a reviewable proposal with the boundary clearly identified. Returning a generic question after every discovery just hands the coordination work back to me.
I also want a conclusion somebody can review without repeating the whole search. A suspicion needs the observations behind it and the reason for any limitation. The agent must be able to end an attempt without taking extra actions simply to manufacture certainty. That exit belongs in the reusable capability itself, alongside the instructions for making progress.
My criterion for turning research into a skill is that the next execution receives a clear scope and returns understandable evidence. The procedure should save explanation on subsequent calls. If I have to remember which access applies and what counts as finished every time, we still have a long prompt waiting for supervision, and that part of the work remains mine.