◉ANTIGRAVITY LABJP
TAG

permission design

4 articles
← Back to all tags
Related:
Antigravity CLI2unattended runs2Antigravity2Claude Code1agent operations1security1agents1indie development1scheduled runs1
⚙ AI Tools/2026-09-15Advanced

The Day I Stopped Splitting by Skill — Three Questions for Routing Work Between Antigravity CLI and Claude Code

An unattended job sat waiting on an approval prompt for three days without a single error line. Here is how I stopped dividing work between two agent CLIs by capability, and started dividing it by where the approval boundary falls.

◈ Agents/2026-07-19Intermediate

The Morning It Wrote Outside the Working Folder: Vetting Every Write Path in One Function

Antigravity CLI 1.1.3 closed a bug where always-proceed mode silently auto-approved writes outside the workspace. Here is a small guard that vets every write path with nothing but path canonicalization and a containment check, plus an escape-attempt test suite.

◈ Agents/2026-07-12Advanced

The Scarier Permission Was Reach, Not Write: Locking an Agent's Outbound Traffic to an Allowlist

When I hand real work to an agent, the thing I guard most isn't file edits, it's where it connects. Here is an 80-line deny-by-default egress gate, plus what 21 nights of unattended runs revealed about the traffic I never saw.

⬡ Integrations/2026-07-11Advanced

Nobody Is There to Say Yes: Writing Unified Permissions as an Unattended Contract

The v2.2.1 unified permission system assumes a person is watching. On a 3 a.m. scheduled run, that assumption quietly breaks. Here is how I declared the policy up front as an allow / deny / queue contract so an unattended agent never stalls on a prompt no one can answer.