v0.2.1: polish, deploy, CI, and literate refactor

- Secret Exposure Gate + Privacy Filter (Bouncer)
- Shell actuator safety harness (timeout, blocked patterns)
- REPL-first enforcement (lisp validation gate, system-prompt-augment)
- Engineering Standards lifecycle (two-track Org-first + REPL-first)
- Literate Programming discipline (one function per block, reflect-back)
- AGENTS.md: thin routing layer, skills are authoritative
- SKILLS_DIR removed, ~/notes fallback eliminated
- opencortex.sh: multi-distro (Debian+Fedora), configure, install service, backup, restore, help
- infrastructure/opencortex.service (systemd user unit)
- Docker: updated to debian:trixie, fixed build context
- GitHub CI: lint + test workflows fixed, trigger on tags only
- Gitea CI: deploy workflow paths fixed
- README: one-line curl install, badges
- USER_MANUAL: Deployment section (bare metal, Docker, backup)
- .gitignore: skills/*.lisp and tests/*.lisp as generated artifacts
- Prose/block refactor across all 35 org files
- Test suite Tier 1: 43/45 pass (env-dependent failures isolated)

skills/org-skill-homoiconic-memory.org

@@ -1,10 +1,10 @@
 #+TITLE: SKILL: Homoiconic Memory (org-skill-homoiconic-memory.org)
 #+AUTHOR: Agent
 #+FILETAGS: :harness:memory:homoiconic:
-#+PROPERTY: header-args:lisp :tangle %%SKILLS_DIR%%/org-skill-homoiconic-memory.lisp
+#+PROPERTY: header-args:lisp :tangle org-skill-homoiconic-memory.lisp
 
 * Overview
-The *Homoiconic Memory* skill provides the capability to treat system memory as executable code and vice-versa.
+Because Lisp is homoiconic (code is data), memory objects can be read as executable forms. This skill provides the bridge between the org-object store and live Lisp evaluation — it can serialize an org-object into an s-expression, evaluate it to reconstruct state, and store the result back as a new object. This is the foundation of the agent's ability to save, restore, and inspect its own cognitive state at runtime.
 
 * Implementation