Tag: agents
The persona pipeline: profile.md to a running container in ten minutes
Edit, PR, merge — the persona restarts with new behavior within ~10 minutes, no human in the loop.
Standing up the persona agent runtime on styra
A persistent container per WIT persona on styra — docker-agent plus shared pgvector memory on an internal-only network.
Git access for agents: why standalone clones beat worktrees
Standalone clones, not git worktree: worktrees share refs — every agent would see (and could clobber) every other agent's branches.
Plumbing week — making the persona → wip → work → main flow actually run
The branch flow makes six parallel agents reviewable rather than chaotic — and it only works if the plumbing doesn't need a plumber.
When your AI team keeps breaking its own git history
The rule 'agents never resolve integration conflicts' is right — but we were applying it to conflicts that needed mechanics, not judgment.
The weekly brief — keeping the overview so the owner doesn't have to
The brief compresses a week of branch archaeology into one page of decisions-needed.
Designing how a #task line becomes a running agent
A poller that detects and routes tasks turns 'read the board, start six sessions' into 'review what landed'.
A map before a mission — consolidating the persona profiles
Coordination without an accurate team map is routing mail to people who don't exist.