When Parallel Agents Ran the Same Task Twice and Quietly Doubled the Bill — Field Notes on Measuring and Stopping Duplicates
The bill for our parallel agents came in about 1.9x higher than expected — because multiple workers were running the same task twice. These are field notes on measuring the duplication, stopping it with idempotency keys, and attributing cost per task.
Protecting Your Agent Stack's Known-Good State with a Single Lockfile — Change-Budget Design for an Era of Simultaneously Moving Parts
When the IDE build, CLI, model, and dependencies all move at once, you can no longer tell which one caused a regression. Here is a change-budget design that pins your known-good state to one lockfile, with working code and operational logs.
When Nobody Reads Your AI Code Reviewer Anymore — Field Notes on Measuring Actioned-Rate
Our production AI code-review agent quietly went hollow over six months. When the team started silently resolving every comment, we instrumented actioned-rate and false-positive rate to bring it back. These are the field notes.
When Streaming Works Locally but Arrives All at Once in Production — Field Notes on Proxy Buffering and Silent Stalls
Stream Gemini through Antigravity over SSE and it flows token-by-token on localhost, then freezes for seconds and dumps the whole answer in production. Field notes on measuring the stall first, then killing proxy buffering, idle disconnects, and reconnect-driven double generation.
Before Your Antigravity Agents Fight Over the Same File — Ownership Manifests and Conflict Detection
Multi-agent workflows do not break at the design stage. They break at runtime. Here are the field notes: an ownership manifest that pins each agent's editable region, a git-only conflict detector, and a three-part handoff contract.
Precedence for Nested AGENTS.md: A Merge Design for Many Projects in One Workspace
Put several projects in one workspace, each with its own AGENTS.md, and which instruction the agent follows turns ambiguous. Root and per-project rules quietly collide; one wins, or both blend. Taking 'closer is stronger' as the base rule, this designs a merge that distinguishes overriding from appending, with working Python and field notes.
Strip Secrets Out of the Agent Logs You Keep: Designing a Redaction Layer
Once you start keeping logs from unattended agents, a token or API key eventually lands in them in plaintext. Rotating the key doesn't unmake the leaked log. This designs a redaction layer that reliably drops secrets right before the write, going beyond regex to register known secrets and mask them for certain, with working Python and field notes.
Scope the MCP Tools You Hand an Agent: A Least-Privilege Allowlist Design
As you add MCP servers to Antigravity 2.0, the set of tools every agent can reach quietly grows into an all-you-can-eat buffet. An agent that only needs to read files seeing delete and deploy tools is an accident waiting to happen. This walks through a least-privilege design that scopes tools per agent role, denies at call time, and gates destructive operations behind a second step, with working Python and field notes.
Your Antigravity Sandbox Isolates Multi-Agents Less Than You Think — Notes on Containing the Blast Radius
An Antigravity sandbox gives you the feeling of isolation, but isolation leaks through three real gaps: shared volumes, over-broad allowed domains, and approval fatigue. Field notes on plugging the leaks, containing the blast radius by design, and proving isolation holds with tests.
Your Antigravity LLM App Drifts on Cost and Quality While the Dashboard Stays Green — Instrumentation Field Notes
Watching only total cost and latency hides the slow drifts that hurt. These are field notes on attributing telemetry by feature, tenant, and prompt version so you catch quality regressions and cost spikes early.
Containing Failure in Antigravity Multi-Agent Systems: Three Boundaries That Stop Cascades
Antigravity multi-agent setups run beautifully in isolation but cascade in production, where one small failure drags the whole orchestration down. These notes organize the fix around three boundaries—layered control, trust separation, and observability with idempotency—down to the TOML and the correlation-ID wrapper.
Delegate the Undoable, Guard the Irreversible — Tiering Agent Autonomy by Reversibility
When you hand production work to an Antigravity agent, the thing that bites first isn't intelligence — it's whether the operation can be undone. Here is a design that sorts every operation into three reversibility tiers and routes each to auto-execution, checkpointed execution, or a human gate, with TypeScript implementations and real numbers from running six apps in parallel.