diff --git a/README.md b/README.md
index e8d8225..4a39221 100644
--- a/README.md
+++ b/README.md
@@ -6,12 +6,12 @@
-**A resident engineer for your support escalations.**
+**A fleet of prebuilt AI teammates for recurring engineering work.**
-Takes a ticket through investigation across your code, telemetry, and live control-plane state. Proposes a fix, waits for human approval, then ships it or drafts the customer reply. Runs against a durable, replayable log — posts evidenced answers, never chats.
+Each one wakes on an event — a pull request, an incident, a deploy — investigates across your code, telemetry, and live control-plane state, opens a scenario-backed fix, and waits for human approval before it ships or drafts the customer reply. Runs against a durable, replayable log — posts evidenced answers, never chats.
[](https://docs.agentsfleet.net)
-[](https://agentsfleet.net)
+[](https://accounts.agentsfleet.net/waitlist)
[](https://opensource.org/licenses/MIT)
diff --git a/index.mdx b/index.mdx
index f6dcae8..fd14ce5 100644
--- a/index.mdx
+++ b/index.mdx
@@ -1,19 +1,19 @@
---
-title: "Your deploy failed. The agent already knows why."
-description: "agentsfleet is an open runtime that owns operational outcomes — an agent wakes on your events, gathers evidence from your own systems, posts a diagnosis with proof, and keeps a replayable log of every move it made."
+title: "A fleet, ready to run."
+description: "agentsfleet is a fleet of prebuilt AI teammates for recurring engineering work — each one wakes on an event, gathers evidence from your own systems, opens a scenario-backed fix, and holds at human approval. Every move is a replayable log."
---
-
- **Stealth-mode testing.** agentsfleet is in private alpha — APIs and agent behavior may change without long deprecation windows. Want a hand calibrating an agent or to join as a design partner? Email [agentsfleet@agentmail.to](mailto:agentsfleet@agentmail.to).
-
+
+ agentsfleet is in early access. Join the waitlist to get in — or email [agentsfleet@agentmail.to](mailto:agentsfleet@agentmail.to) to come on as a design partner.
+
-## The 2am problem
+## A teammate that owns the outcome
-A deploy fails at 2am. Your on-call engineer is asleep, so someone guesses from a stack trace and ships a patch on hope. By morning the context is gone — no record of what was tried, evidence scattered across CI, logs, dashboards, and chat.
+A pull request lands. A deploy fails at 2am. Normally someone guesses from a stack trace and ships a patch on hope — and by morning the context is gone, evidence scattered across CI, logs, dashboards, and chat.
-agentsfleet owns the outcome instead. An **agent** is a durable runtime described in two markdown files — `SKILL.md` (the senior engineer's playbook in prose) plus `TRIGGER.md` (which tools the agent can reach and which events wake it). When a trigger fires, the agent gathers the evidence, reasons over it, posts an evidenced diagnosis to your Slack channel, and keeps a replayable log of every move it made.
+A teammate owns the outcome instead. An **agent** is a durable runtime described in two markdown files — `SKILL.md` (the senior engineer's playbook in prose) plus `TRIGGER.md` (which tools the agent can reach and which events wake it). When a trigger fires, it gathers the evidence, reasons over it, opens a scenario-backed fix, posts an evidenced diagnosis to your Slack channel, and keeps a replayable log of every move — holding at human approval before anything merges or ships.
-The flagship `platform-ops` wakes on a GitHub Actions deploy failure, pulls workflow logs + your hosting provider's app state + your data-plane's health, and posts a one-paragraph diagnosis with a leading hypothesis. The same agent is reachable via `agentsfleet steer` for a manual investigation. The runtime is open source, you bring your own model and pay your provider directly, and the agent never sees your raw keys — only `${secrets.NAME.FIELD}` placeholders, swapped for real values outside the sandbox.
+`platform-ops` is **one** prebuilt teammate, and more keep joining the fleet — code review, incident triage, security review. It wakes on a GitHub Actions deploy failure, pulls workflow logs + your hosting provider's app state + your data-plane's health, and posts a one-paragraph diagnosis with a leading hypothesis. The same agent is reachable via `agentsfleet steer` for a manual investigation. The runtime is open source, you bring your own model and pay your provider directly, and the agent never sees your raw keys — only `${secrets.NAME.FIELD}` placeholders, swapped for real values outside the sandbox.
## Get started