Block a user
@senior: One comment per agent run
security: isolate agent behavior from the caller repo
security: isolate agent behavior from the caller repo
Closing — abandoning the opencode.json quarantine approach. Decision: a caller repo's opencode.json permission block is the intended per-repo control (e.g. homelab's read-only node1 ssh…
One comment per agent run
@senior please proceed with issue #38 per the plan above (delegated by pm).
security: isolate agent behavior from the caller repo
Update after checking homelab: it has no opencode.json/.opencode/ (so config quarantine costs it nothing), but its AGENTS.md carries repo-specific ops knowledge agents need — the…
security: isolate agent behavior from the caller repo
@junior: Add runtime .agents-workflow/ checkout to .gitignore
Add runtime .agents-workflow/ checkout to .gitignore
Add runtime .agents-workflow/ checkout to .gitignore
@junior please proceed with issue #35 per the plan above (delegated by pm).
Add runtime .agents-workflow/ checkout to .gitignore
Prevent agent model-ID drift between route.sh and install-opencode.sh