diff --git a/hive-ag3nt/prompts/agent.md b/hive-ag3nt/prompts/agent.md
index 1f3d0330..46a1eddc 100644
--- a/hive-ag3nt/prompts/agent.md
+++ b/hive-ag3nt/prompts/agent.md
@@ -15,6 +15,6 @@ Claude session (OAuth credentials) lives at `/root/.claude/` and persists across
 
 **Shared space**: `/shared` is accessible to all agents (read/write). Only put things here you're willing to lose — other agents may delete them. Use for explicit cross-agent communication or shared artifacts when appropriate.
 
-Keep messages short — a few sentences each. For anything big (file listings, long diffs, transcripts, analysis): write the payload to `/agents/{label}/state/<descriptive-name>` and `send` a short pointer ("dropped the cluster audit in /agents/{label}/state/cluster-audit-2026-05.md, headline: 3 nodes over 80% mem"). The manager + operator can read your state from the host as `/agents/{label}/state/`. Sub-agent peers can't read each other's state directly — go through the manager if a payload needs to reach another sub-agent.
+Keep messages short — a few sentences each. For anything big (file listings, long diffs, transcripts, analysis): write the payload to `/agents/{label}/state/<descriptive-name>` and `send` a short pointer ("dropped the cluster audit in /state/cluster-audit-2026-05.md, headline: 3 nodes over 80% mem"). The manager + operator can read your state from the host as `/agents/{label}/state/`. Sub-agent peers can't read each other's `/state/` directly — go through the manager if a payload needs to reach another sub-agent.
 
 When your inbox has a message, handle it and stop. Don't narrate intent — act.
diff --git a/hive-ag3nt/prompts/manager.md b/hive-ag3nt/prompts/manager.md
index f65e106f..c08f5b8a 100644
--- a/hive-ag3nt/prompts/manager.md
+++ b/hive-ag3nt/prompts/manager.md
@@ -79,7 +79,7 @@ Durable knowledge:
 
 Keep messages short — a few sentences each. For anything big (digests, agent rosters, plans, transcripts) write the payload to a file and `send` a short pointer:
 
-- To a sub-agent X: write to `/agents/X/state/<descriptive-name>` and tell them "see /agents/X/state/<descriptive-name>".
+- To a sub-agent X: write to `/agents/X/state/<descriptive-name>` and tell them "see /state/<descriptive-name>".
 - To the operator: write to your own `/state/<descriptive-name>` (host path `/var/lib/hyperhive/agents/hm1nd/state/`) and tell them where to look.
 - For shared artifacts (coordination, common reference data): write to `/shared/<descriptive-name>`. Only put things here you're willing to lose — other agents may delete them.
 
diff --git a/hive-ag3nt/src/plugins.rs b/hive-ag3nt/src/plugins.rs
index 00254f9f..f1302d00 100644
--- a/hive-ag3nt/src/plugins.rs
+++ b/hive-ag3nt/src/plugins.rs
@@ -6,10 +6,6 @@
 //! is expected to be idempotent so reinstalling on each container
 //! recreate is fine. Failures log a warning but do not abort boot —
 //! we'd rather start without a plugin than refuse to serve.
-//!
-//! Before installing, all configured marketplaces are updated so that
-//! plugin specs resolve against current index data. Marketplace update
-//! failures are non-fatal — stale index is better than no install attempt.
 
 use std::path::Path;
 
@@ -19,30 +15,6 @@ use crate::client;
 
 const PLUGINS_PATH: &str = "/etc/hyperhive/claude-plugins.json";
 
-/// Update all configured plugin marketplaces. Non-fatal — logs a warning
-/// on failure but does not abort the install sequence.
-async fn update_marketplaces() {
-    match Command::new("claude")
-        .args(["plugin", "marketplace", "update"])
-        .output()
-        .await
-    {
-        Ok(out) if out.status.success() => {
-            tracing::info!("claude plugin marketplace update ok");
-        }
-        Ok(out) => {
-            tracing::warn!(
-                status = ?out.status,
-                stderr = %String::from_utf8_lossy(&out.stderr),
-                "claude plugin marketplace update failed (non-fatal)",
-            );
-        }
-        Err(e) => {
-            tracing::warn!(error = ?e, "claude plugin marketplace update spawn failed (non-fatal)");
-        }
-    }
-}
-
 /// Install every plugin in `/etc/hyperhive/claude-plugins.json`. When
 /// `notify_recipient` is `Some(name)`, install failures also get sent
 /// as a hyperhive message to that recipient (typically `"manager"` for
@@ -61,10 +33,6 @@ pub async fn install_configured(socket: &Path, notify_recipient: Option<&str>) {
             return;
         }
     };
-    if specs.is_empty() {
-        return;
-    }
-    update_marketplaces().await;
     for spec in specs {
         match Command::new("claude")
             .args(["plugin", "install", &spec])