docs(machine-readable): update 6a2 for 2026-04-11 Fly cold-start hardening session

- STATE.a2ml: v2.1.1, add session-2026-04-11 with Fly health check +
  90s grace + curl Containerfile changes + rationale (Isabelle heap / Julia JIT)
- ECOSYSTEM.a2ml: v0.2.0, full content (was nearly empty) — purpose, role,
  related projects, deployment config including Fly app URL and cold-start note

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Author: hyperpolymath

.machine_readable/6a2/ECOSYSTEM.a2ml

@@ -1,10 +1,40 @@
 # SPDX-License-Identifier: PMPL-1.0-or-later
-# ECOSYSTEM.a2ml — Ecosystem position
-# Converted from ECOSYSTEM.scm on 2026-03-15
+# Copyright (c) 2026 Jonathan D.A. Jewell (hyperpolymath) <j.d.a.jewell@open.ac.uk>
+#
+# ECOSYSTEM.a2ml — Ecosystem position for echidna
 
 [metadata]
+version = "0.2.0"
+last-updated = "2026-04-11"
 project = "echidna"
 ecosystem = "hyperpolymath"
 
+[project]
+name = "ECHIDNA"
+full-name = "Extensible Cognitive Hybrid Intelligence for Deductive Neural Assistance"
+purpose = """Trust-hardened neurosymbolic theorem proving platform. 48 prover backends
+(Agda, Lean4, Isabelle, Coq, Idris2, F*, Z3, CVC5, Vampire, and 39 more). Sandboxed
+solver execution (Podman/bubblewrap), proof certificate checking, axiom tracking,
+GNN-guided proof search, Julia ML inference layer."""
+role = "prover-dispatcher"
+
 [position]
+tier = "core-infrastructure"
 type = "component"
+
+[related-projects]
+projects = [
+  { name = "proven-servers (nesy-solver-api)", relationship = "consumer", notes = "nesy-solver-api calls /api/verify on echidna-nesy.flycast:8090. Cold-start grace 90s." },
+  { name = "hypatia", relationship = "sibling-neurosym", notes = "Hypatia performs neurosymbolic CI scanning; echidna does theorem proving. Both are Tier 2 formal verification tools." },
+  { name = "verisimdb", relationship = "planned-integration", notes = "Proof attempt results should feed VeriSimDB for longitudinal tracking." },
+  { name = "neural-foundations", relationship = "monorepo-satellite", notes = "echidna is mirrored as a submodule/satellite inside neural-foundations." },
+  { name = "maa-framework", relationship = "sibling", notes = "Absolute-zero and related MA/ML tools in same neurosymbolic tier." },
+]
+
+[deployment]
+fly-app = "echidna-nesy"
+region = "lhr"
+url = "http://echidna-nesy.flycast:8090 (private 6PN — auto-wake via .flycast)"
+health-endpoint = "/api/health"
+grace-period = "90s"
+cold-start-note = "Isabelle Main heap ~1-2GB maps at first use; Julia JIT runs on first inference. Minimum 90s grace."

.machine_readable/6a2/STATE.a2ml

@@ -4,8 +4,8 @@
 
 [metadata]
 project = "echidna"
-version = "2.0.0"
-last-updated = "2026-04-05"
+version = "2.1.1"
+last-updated = "2026-04-11"
 status = "active"
 
 [project-context]
@@ -73,3 +73,13 @@ changes = [
 ]
 test-counts = { rust = 765, julia = 63, shell-checks = 68, benchmarks = 13, isabelle-module = 13 }
 stub-audit-result = "Only Isabelle was truly stubbed; all other 47 external-solver backends spawn real subprocesses. 2 backends (metamath, typed_wasm) are intentionally pure-Rust in-process verifiers."
+
+[session-2026-04-11]
+summary = "Fly.io cold-start hardening for echidna-nesy deployment"
+changes = [
+  "fly.toml: [[services.tcp_checks]] → [[services.http_checks]] on /api/health; grace_period 30s → 90s; timeout 5s → 10s; added comment explaining Isabelle heap + Julia JIT warm-up requirement",
+  ".containerization/Containerfile.full: added curl to runtime apt-get layer; HEALTHCHECK switched from --version flag to curl -fs http://localhost:8090/api/health; --start-period 5s → 90s",
+  "Both echidna standalone and neural-foundations satellite copies updated identically",
+  "server.v (proven-nesy-solver-api): call_echidna_verify read_timeout raised to 90s to tolerate echidna cold-start",
+]
+rationale = "Isabelle Main heap maps ~1-2GB at first use; Julia JIT runs on first inference call. Without 90s grace the Fly health probe would kill the machine before it was ready, causing cascading restart loops."