Home / Articles / Governed semi-autonomous development, read in a newsroom run by agents / Versions / v1.1.0
Governed semi-autonomous development, read in a newsroom run by agents: what changed in v1.1.0
From v1.0.0 (2026-10-21, ce2e11671) to v1.1.0 (2026-10-27, db67c5ce1), paragraph by paragraph.
0 paragraphs added, 0 removed, 3 changed in place, 42 unchanged. About 0 words added and 82 removed. Insertions are marked like this, deletions like this; unchanged runs are folded to one line; figures appear as their file names.
2 unchanged paragraphs
!shot gs-map.webp | images/ | The same ideas, arrived at separately: the method's principles on the left, principles from Governed semi-autonomous development; on the right, where each shows up in whatour we have built and published, from the inbox agent that drafts while a person sends, to one check in 57 out of the agent's reach.work. | comparison
21 unchanged paragraphs, under In short {#in-short}, A method born from practice {#method}, The same ideas, arrived at separately {#same}…
!shot gs-self.webp | images/ | This newsroom measured against the method's Lite profile, layer by layer:layer, infrastructure,with sessions,the messagesshortfalls and state meet it; tools and process partly; permissions fall short by athe recorded choice, because the agent publishes; and repository controls fall short, because the agent pushes to the release branch and can edit the CI that gates it.choice. | comparison
6 unchanged paragraphs, under What we would add {#add}
!shot gs-add.webp | images/ | Five suggestions:suggestions, from a column per rule saying what keeps it;it behaviour policies for the agents inside the product; state in encrypted vaults; the owner measures he proposes, measured; andto memory loss as a design input. | concept
13 unchanged paragraphs, under What this does not claim {#limits}, Where this comes from {#sources}