repo-nexus 0.7.1 → 0.7.2

package/lib/commands/sync.sh

@@ -108,6 +108,9 @@ cmd_sync() {
     [ "$_sync_quiet" -eq 0 ] && log_ok "Upgraded workspace configuration and routing instructions to v$RNEX_VERSION"
   fi
 
+  # Dispatch before_sync plugin lifecycle hook
+  run_hook "before_sync" "$@"
+
   # 1. Ensure .gitignore ignores repos/, local configs; ensure .ignore has !repos/*
   reconcile_gitignore "$NEXUS_DIR"
 
@@ -184,6 +187,9 @@ cmd_sync() {
   # 10. Update sync stamp
   update_sync_stamp "$NEXUS_DIR"
 
+  # Dispatch after_sync plugin lifecycle hook
+  run_hook "after_sync" "$@"
+
   if [ "$_sync_quiet" -eq 0 ]; then
     log_ok "Sync complete — workspace, plugins, prompts, and member .rnex directories reconciled"
   fi

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "repo-nexus",
-  "version": "0.7.1",
+  "version": "0.7.2",
   "description": "Lightweight, zero-dependency Virtual Meta-Repo companion for multi-repo workspaces and shared AI context (no git submodules)",
   "bin": {
     "repo-nexus": "rnex",

package/toolkit/plugins/karpathy-llm/hooks/after_sync.sh

@@ -88,8 +88,8 @@ RAW_DIR_VAL=$(grep -h -E '^\s*raw_dir:' "$RNEX_ROOT/rnex.yaml" "$RNEX_ROOT/.loca
 [ -z "$RAW_DIR_VAL" ] && RAW_DIR_VAL="$(get_cfg "raw_dir" "./docs")"
 RAW_DIR="${RAW_DIR_VAL:-./docs}"
 
-# Clear and sync the custom local raw folder files to raw intake (if any exist)
-if [ -d "$RNEX_ROOT/$RAW_DIR" ]; then
+# Clear and sync custom local raw folder files to raw intake (skip if RAW_DIR is a documentation site)
+if [ -d "$RNEX_ROOT/$RAW_DIR" ] && [ ! -d "$RNEX_ROOT/$RAW_DIR/.vitepress" ]; then
   _raw_abs="$(cd "$RNEX_ROOT/$RAW_DIR" 2>/dev/null && pwd -P || true)"
   _wiki_raw_abs="$(cd "$WIKI_RAW_DIR" 2>/dev/null && pwd -P || true)"
   if [ -n "$_raw_abs" ] && [ "$_raw_abs" != "$_wiki_raw_abs" ]; then
@@ -113,11 +113,33 @@ if [ -d "$RNEX_ROOT/$REPOS_DIR" ]; then
     [ ! -d "$repo_path" ] && continue
     repo_name=$(basename "$repo_path")
     
-    # Ignore repos that are disabled locally (checking status state)
-    # Checks if enabled: false exists for this repo in configuration
-    if grep -A5 "  $repo_name:" "$RNEX_ROOT/rnex.yaml" "$RNEX_ROOT/.local.rnex.yaml" 2>/dev/null | grep -q "enabled: false"; then
-      continue
-    fi
+    # Check if this repository is explicitly disabled in .local.rnex.yaml or rnex.yaml (block-scoped)
+    _is_disabled=""
+    for _cfg_file in "$RNEX_ROOT/.local.rnex.yaml" "$RNEX_ROOT/rnex.yaml"; do
+      [ -f "$_cfg_file" ] || continue
+      _en="$(awk -v repo="$repo_name" '
+        /^repos:[ ]*$/ || /^repos:[ ]*#/ { r=1; next }
+        r && /^[^ #]/ { r=0 }
+        r && match($0, "^  " repo ":[ ]*(#.*)?$") { in_target=1; next }
+        in_target && /^  [^ #]/ { in_target=0 }
+        in_target && match($0, "^    enabled:[ ]*") {
+          line = $0
+          sub("^    enabled:[ ]*", "", line)
+          sub(/[ ]*#.*$/, "", line)
+          gsub(/^["\x27]|["\x27]$/, "", line)
+          print line
+          exit
+        }
+      ' "$_cfg_file")"
+      if [ -n "$_en" ]; then
+        case "$(printf '%s' "$_en" | tr '[:upper:]' '[:lower:]')" in
+          false|no|0|disabled|off) _is_disabled=1 ;;
+          *) _is_disabled=0 ;;
+        esac
+        break
+      fi
+    done
+    [ "$_is_disabled" = "1" ] && continue
 
     # Read and sync any README file with case-insensitive check
     for readme_name in README.md readme.md README.markdown README; do

package/toolkit/prompts/index.md

@@ -13,10 +13,11 @@ When an AI assistant is assigned a cross-repo or workspace-wide task, consult th
    - **Purpose**: Cross-repository health, git status, branch tracking, and configuration validation.
    - **When to use**: Auditing workspace status, verifying uncommitted changes, or syncing member repositories before releases.
 
-3. **[`rnex-ingest.md`](./rnex-ingest.md)** (alias: [`rnex-wiki-ingest.md`](./rnex-wiki-ingest.md))
+3. **[`rnex-wiki-ingest.md`](./rnex-wiki-ingest.md)**
    - **Purpose**: Autonomous ingestion of intake materials from configured raw intake directory into atomic Karpathy LLM Wiki pages.
    - **When to use**: Decomposing research papers, documentation, transcripts, or specifications into the knowledge base.
 
-4. **[`rnex-lint.md`](./rnex-lint.md)** (alias: [`rnex-wiki-lint.md`](./rnex-wiki-lint.md))
+4. **[`rnex-wiki-lint.md`](./rnex-wiki-lint.md)**
    - **Purpose**: Deterministic graph audit, dead link validation, `mentioned_in` backlink reconciliation, and index rebuild.
    - **When to use**: Routine wiki maintenance, pre-commit validation, or after adding/renaming wiki pages.
+

package/toolkit/prompts/rnex-cross-repo-feature.md

@@ -1,35 +1,16 @@
-# Cross-Repository Feature Implementation Workflow
+# Operational Task: Cross-Repository Feature Implementation
 
-Use this prompt to guide the planning, execution, and verification of features spanning multiple member repositories in a Repo Nexus workspace.
+You are tasked with planning, coordinating, and executing a feature spanning multiple member repositories in this Repo Nexus workspace.
 
----
+## Context & Inputs
+- Target Scope: `repos/<name>/` directory trees
+- Workspace Manifest: `rnex.yaml` and `.local.rnex.yaml`
+- AI Instructions: `.rnex/ai-instructions.md`
 
-## 1. Discovery & Scope Analysis
-1. Inspect `.local.rnex.yaml` (if present) and `rnex.yaml` to identify active member repositories in `repos/<name>/`.
-2. Inspect `repos/<name>/.rnex/` in each target repository for repo-specific architectural rules or constraints.
-3. If `wiki/` is present, read `wiki/hot.md` and `wiki/index.md` to orient on existing domain contracts and data models.
-4. Clearly define the interface contract or boundary change between repositories (e.g. API payload, shared type definitions, database schema).
+## Skill & Workflow Router
+1. Read the operational skill requirements in `.rnex/skills/rnex-cross-repo-feature/SKILL.md`.
+2. Follow the staged execution protocol strictly in `.rnex/workflows/rnex-cross-repo-feature.md`.
+3. Adhere to mandatory session startup, routing protocols, and member repo boundary rules in `.rnex/ai-instructions.md`.
+4. Do NOT execute uncompiled heuristics or perform unguided exploratory codebase scans.
 
-## 2. Staged Implementation Plan
-1. Order repository changes by dependency hierarchy (e.g. backend / shared models first, frontend / consumers second).
-2. For each member repository:
-   - Identify specific files to modify.
-   - Check member repo test suite requirements.
-
-## 3. Execution
-1. Perform file modifications directly in the target `repos/<name>/` directory trees.
-2. Maintain documentation and type integrity across boundary contracts.
-3. Verify changes locally within each modified repository using `rnex exec` or targeted test runners:
-   ```sh
-   rnex exec npm test
-   # or targeted:
-   cd repos/<name> && npm test
-   ```
-
-## 4. Git & Commit Hygiene
-1. Member repositories are autonomous Git repositories:
-   - Commit changes independently within each member repository root:
-     ```sh
-     cd repos/<name> && git add -A && git commit -m "feat: <description>"
-     ```
-   - Never commit member repository source code directly to the Repo Nexus workspace root.
+Execute the referenced workflow across affected member repositories. Maintain boundary contracts, run targeted test suites, and commit changes independently within each member repository root.

package/toolkit/prompts/rnex-wiki-ingest.md

@@ -1,14 +1,18 @@
-# Operational Task: Ingest New Material
+# Operational Task: Karpathy LLM Wiki Ingestion
 
-You are tasked with executing the ingestion and atomic break-down workflow.
+You are tasked with executing the knowledge ingestion and atomic decomposition workflow for the Karpathy LLM Wiki.
 
 ## Context & Inputs
-- Source files location: `.rnex/wiki/raw/` (inclusive of active member repo `README_*.md` files)
-- Goal: Deconstruct files into individual interlinked pages in `.rnex/wiki/`.
+- Target Scope: Curated wiki directory (`wiki_dir`, default: `.rnex/wiki/`)
+- Intake Sources: Configured raw intake directory (`raw_dir`, default: `./docs`) and synchronized member repository READMEs
+- Workspace Manifest: `rnex.yaml` and `.local.rnex.yaml`
+- AI Instructions: `.rnex/ai-instructions.md`
 
-## Workflow & Guidelines Router
-1. Read the strict ingest steps and verification metrics only from `.rnex/plugins/karpathy-llm/workflows/wiki-ingest.md`.
-2. Follow styling, structural, and behavioral constraints inside `.rnex/plugins/karpathy-llm/rules/behavioral.md`.
-3. Adhere to global instruction patterns in `.rnex/ai-instructions.md`.
+## Skill & Workflow Router
+1. Read the operational skill requirements in `.rnex/plugins/karpathy-llm/skills/rnex-wiki-ingest/SKILL.md`.
+2. Follow the atomic decomposition and ingestion protocol strictly in `.rnex/plugins/karpathy-llm/workflows/rnex-wiki-ingest.md`.
+3. Follow architectural and behavioral constraints in `.rnex/plugins/karpathy-llm/rules/rnex-behavioral.md`.
+4. Adhere to mandatory session startup, routing protocols, and global rules in `.rnex/ai-instructions.md`.
+5. Do NOT invent new frontmatter schemas, skip atomic decomposition, or alter source materials directly.
 
-Transform unmodified raw documents on ingestion following exactly the referenced workflow. Do not use inline logic or generate outside explicit instructions. Run 'wiki-lint' upon completion to review accuracy. Update Operation Log (`.rnex/wiki/_log.md`).
+Decompose intake documents into atomic, interlinked wiki pages with typed frontmatter relations. Update the master index and append a timestamped entry to the operation log.

package/toolkit/prompts/rnex-wiki-lint.md

@@ -1,15 +1,17 @@
-# Operational Task: Workspace Wiki Lint & Audit
+# Operational Task: Karpathy LLM Wiki Lint & Graph Audit
 
-You are tasked with executing the Workspace Wiki Lint workflow.
+You are tasked with auditing graph integrity, validating bidirectional relations, and repairing links for the Karpathy LLM Wiki.
 
 ## Context & Inputs
-- Target Scope: `.rnex/wiki/` (including `.rnex/wiki/raw/` and synchronized member READMEs)
-- Custom Rules & Workflows loaded: `.rnex/plugins/karpathy-llm/`
-- Current Workspace Manifest: `rnex.yaml` and `.local.rnex.yaml`
+- Target Scope: Curated wiki directory (`wiki_dir`, default: `.rnex/wiki/`)
+- Workspace Manifest: `rnex.yaml` and `.local.rnex.yaml`
+- AI Instructions: `.rnex/ai-instructions.md`
 
-## Verification Rule Router
-1. Locate and parse your validation instructions exclusively from `.rnex/plugins/karpathy-llm/workflows/wiki-lint.md`.
-2. Evaluate adherence based on the agent behavior standards described in `.rnex/plugins/karpathy-llm/rules/behavioral.md` and compiled rules in `.rnex/ai-instructions.md`.
-3. Do NOT invent new lint checks or execute uncompiled heuristic instructions.
+## Skill & Workflow Router
+1. Read the operational skill requirements in `.rnex/plugins/karpathy-llm/skills/rnex-wiki-lint/SKILL.md`.
+2. Follow the multi-step audit and graph reconciliation protocol in `.rnex/plugins/karpathy-llm/workflows/rnex-wiki-lint.md`.
+3. Follow architectural and behavioral constraints in `.rnex/plugins/karpathy-llm/rules/rnex-behavioral.md`.
+4. Adhere to mandatory session startup, routing protocols, and global rules in `.rnex/ai-instructions.md`.
+5. Do NOT invent new lint checks or execute uncompiled heuristic instructions.
 
-Execute the 10-step protocol defined strictly within the referenced workflow. Audit `raw_dir` items, link paths, index categorization, and relational YAML frontmatter. Provide your structured audit and write feedback output to the Audit trail in `.rnex/wiki/_log.md`.
+Execute the referenced lint workflow across all wiki pages. Validate frontmatter relation filepaths, recompute `mentioned_in` backlinks, identify orphan pages, reconcile the master index, and record findings in the operation log.

package/toolkit/prompts/rnex-workspace-audit.md

@@ -1,34 +1,16 @@
-# Repo Nexus Workspace Audit Workflow
+# Operational Task: Workspace Configuration & Health Audit
 
-Use this prompt to audit workspace configuration integrity, member repository git statuses, and active plugin states.
+You are tasked with executing a comprehensive audit of the Repo Nexus workspace configuration, member repository git statuses, and active plugin states.
 
----
+## Context & Inputs
+- Target Scope: Root workspace configuration, `.rnex/`, and active member repositories in `repos/<name>/`
+- Workspace Manifest: `rnex.yaml` and `.local.rnex.yaml`
+- AI Instructions: `.rnex/ai-instructions.md`
 
-## 1. Configuration & Health Audit
-1. Run `rnex status` to inspect:
-   - Configuration files (`rnex.yaml` and `.local.rnex.yaml`).
-   - Main AI instructions file existence and routing markers.
-   - VS Code / Cursor `.code-workspace` synchronization.
-   - Git ignore status for `repos/`, `.rnex/`, and local overrides.
-   - Active plugins and member repository registry.
+## Skill & Workflow Router
+1. Read the operational skill requirements in `.rnex/skills/rnex-workspace-audit/SKILL.md`.
+2. Follow the multi-step audit and remediation protocol in `.rnex/workflows/rnex-workspace-audit.md`.
+3. Adhere to mandatory session startup, routing protocols, and member repo boundary rules in `.rnex/ai-instructions.md`.
+4. Do NOT execute uncompiled heuristics or alter workspace configuration outside the referenced workflow.
 
-## 2. Multi-Repository Git Status Audit
-1. Check working tree cleanliness across all active member repositories:
-   ```sh
-   rnex exec git status -s
-   ```
-2. Check branch tracking and unpushed/unpulled commits:
-   ```sh
-   rnex exec git status -uno
-   ```
-3. Report any detached HEADs, uncommitted modifications, or untracked changes.
-
-## 3. Reconciliation & Remediation
-1. If any configuration issues, missing directories, or unlinked plugin assets are detected, run:
-   ```sh
-   rnex status --fix
-   ```
-2. If member repositories declared in `rnex.yaml` are missing on disk, run:
-   ```sh
-   rnex clone
-   ```
+Execute the audit according to the referenced workflow and report configuration integrity, working tree cleanliness, and active plugin health.

package/toolkit/plugins/karpathy-llm/rules/behavioral.md

@@ -1,47 +0,0 @@
-# Karpathy Coding Agent Behavioral Principles
-
-> "The delicate art and science of context engineering: filling the context window with just the right information for the next step." — Andrej Karpathy
-
-These rules define standing behavioral principles for AI coding assistants (Cursor, Claude Code, GitHub Copilot, Antigravity, Windsurf) working in this workspace.
-
----
-
-## 1. The Four Cardinal Principles
-
-### Principle 1: Think Before Coding
-* **State Assumptions Explicitly:** Before implementing any non-trivial change, state your understanding of the problem, assumptions made, and proposed solution.
-* **Surface Trade-Offs:** If multiple approaches exist, outline the alternatives with their trade-offs before generating code.
-* **Verify Repository Boundaries:** In multi-repo workspaces, determine which member repositories are affected. Never make silent architectural assumptions that span across repo boundaries.
-
-### Principle 2: Simplicity First
-* **Favor Readable Over Clever:** Write the simplest code that solves the immediate problem. Avoid unnecessary abstractions, speculative generics, and premature optimizations.
-* **Zero Boilerplate Bloat:** Do not introduce extra wrapper layers, helper classes, or framework-heavy patterns unless explicitly requested or required by existing codebase conventions.
-* **Less Code is Better Code:** If an existing utility, POSIX primitive, or library function does the job, reuse it.
-
-### Principle 3: Surgical Changes
-* **Minimal Blast Radius:** Touch *only* the lines and files required to achieve the goal.
-* **Preserve Documentation & Comments:** Never delete or overwrite existing comments, documentation, or code structure unrelated to your changes.
-* **Zero Unintended Reformatting:** Avoid running blanket code formatters across entire files that create massive, unreadable git diffs. Keep diffs tight, focused, and reviewable.
-
-### Principle 4: Goal-Driven Execution & Verification
-* **Define Success Criteria Upfront:** Know how you will prove that your change works before you write the first line of code.
-* **Verify with Automated Checks:** Run linters, unit tests, and build commands before declaring a task complete.
-* **Inspect the Diff:** Review your own git diff before committing or presenting the solution to ensure no extraneous changes were introduced.
-
----
-
-## 2. Multi-Repo Context Engineering
-
-In a Repo Nexus workspace, independent repositories are organized into `./repos/<name>/` as autonomous Git repositories. Coding agents must follow these rules:
-
-1. **Repository Boundaries:**
-   - Edits made to `repos/<name>/` modify the underlying member repository directly.
-   - Member repositories are autonomous projects. Do not introduce cross-repository source imports or shared runtime dependencies unless an explicit monorepo architecture is configured.
-
-2. **Member .rnex Directory Routing:**
-   - Member repositories may contain an `.rnex/` directory (e.g. `repos/<name>/.rnex/`) containing repository-specific documents, instructions, rules, workflows, and scripts. Coding assistants must inspect this directory for member-specific instructions.
-   - Run git operations (commits, branches, pushes) directly within the respective member repository root (`repos/<name>/`).
-
-3. **Context Economy:**
-   - Do not redundantly read entire files or directory trees when targeted symbol lookups or line ranges suffice.
-   - Rely on active workspace context, indexing, and compiled documentation rather than re-scanning raw files repeatedly.

package/toolkit/plugins/karpathy-llm/workflows/wiki-ingest.md

@@ -1,66 +0,0 @@
-# Wiki Ingestion Workflow (`wiki-ingest`)
-
-Universal workflow for decomposing and ingesting raw source materials into atomic Karpathy LLM Wiki pages.
-
----
-
-## Ingestion Protocol
-
-1. **Resolve Configuration & Scan Intake:**
-   - Read `rnex.yaml` (and `.local.rnex.yaml` if present) to resolve the active `raw_dir` (default: `./docs`) and `wiki_dir` (default: `.rnex/wiki`) under `plugins.karpathy-llm`.
-   - Scan `<raw_dir>/` (default: `./docs`) for source materials (articles, transcripts, documents, research notes, meeting logs) not yet recorded in `<wiki_dir>/_log.md` (default: `.rnex/wiki/_log.md`).
-   - All source material must reside in `<raw_dir>/` unmodified. Never write directly to `<wiki_dir>/` without ingesting.
-
-2. **Read & Clarify:**
-   - Read each source fully.
-   - If scope, domain boundaries, target audience, or level of detail is unclear, ask one round of clarifying questions before proceeding.
-
-3. **Atomic Decomposition:**
-   - Decompose each source into atomic wiki pages — **one per distinct concept, person, organization, event, or theme** (or architecture decision/contract).
-   - A single source document commonly produces **5–25 pages**. Never collapse an entire source into a single file.
-   - Place pages according to workspace wiki organization:
-     - **Flat:** directly under `<wiki_dir>/` (e.g., `<wiki_dir>/<topic>.md`)
-     - **Structured:** in categorical subfolders within `<wiki_dir>/`:
-       - Research → `concepts/`, `people/`, `organizations/`, `sources/`, `analysis/`
-       - Second brain → `projects/`, `people/`, `decisions/`, `logs/`
-       - Content archive → `sources/`, `people/`, `tools/`, `concepts/`
-       - Codebase → `architecture/`, `decisions/`, `runbooks/`, `people/`
-
-4. **Frontmatter Relations:**
-   - For each page, write the required YAML frontmatter with typed relations:
-   ```yaml
-   ---
-   title: ""
-   tags: []
-   last_updated: YYYY-MM-DD
-   # Typed relation fields — paths relative to <wiki_dir>/ (or workspace root for raw sources)
-   sources: []          # <raw_dir>/ files this page was derived from
-   related: []          # thematically related wiki pages
-   extends: []          # pages this one builds upon or specialises
-   contradicts: []      # pages with conflicting information
-   mentioned_in: []     # pages that link to this one (leave empty; maintained by lint)
-   ---
-   ```
-   - **Crucial:** The relation graph lives in frontmatter, not in prose links. Agents traverse the graph by reading frontmatter fields, not by scanning body text.
-
-5. **Author Concise Body:**
-   - Write a concise body with summary, key facts, and insights.
-   - Add inline standard Markdown links (`[Label](./path.md)`) where contextually useful in prose.
-
-6. **Update Navigation Index (Annotated Links):**
-   - Update `<wiki_dir>/index.md` (default: `.rnex/wiki/index.md`) with links to all new pages in their corresponding categorized sections.
-   - Include a concise 1-line summary next to each link (e.g. `- [Concept Name](path.md): One-line summary of role or contract`) so agents pinpoint target pages in a single read without exploratory scanning.
-
-7. **Update Hot Cache (Bounded ~500-Word Ceiling):**
-   - Update `<wiki_dir>/hot.md` (default: `.rnex/wiki/hot.md`) with the rolling orientation summary across 3 structured sections:
-     1. **Architecture & Member Repo Map**: Active domain ownership per member repo.
-     2. **Current Focus & Recent Ingests**: Active topics from recent ingests.
-     3. **Key Contracts & Boundaries**: Critical cross-repo or subsystem rules.
-   - **Enforce Token Economy**: Keep `<wiki_dir>/hot.md` strictly under ~500 words. Condense or prune older focus items to prevent bloat.
-
-8. **Append Operation Log:**
-   - Append a timestamped entry to `<wiki_dir>/_log.md` (default: `.rnex/wiki/_log.md`) detailing:
-     - Date & time
-     - Raw intake file processed
-     - List of atomic pages generated
-     - Any flagged follow-ups or knowledge gaps

package/toolkit/plugins/karpathy-llm/workflows/wiki-lint.md

@@ -1,50 +0,0 @@
-# Wiki Lint & Maintenance Workflow (`wiki-lint`)
-
-Periodic health check and maintenance workflow to preserve wiki graph integrity, validate typed relations, recompute bidirectional links, resolve knowledge gaps, and rebuild indexes.
-
----
-
-## Lint & Maintenance Protocol
-
-1. **Resolve Configuration & Verify Paths:**
-   - Read `rnex.yaml` (and `.local.rnex.yaml` if present) to resolve the active `wiki_dir` (default: `.rnex/wiki`) and `raw_dir` (default: `./docs`) under `plugins.karpathy-llm`.
-   - For every page under `<wiki_dir>/`, verify that all paths in frontmatter relation fields (`related`, `extends`, `contradicts`, `sources`) resolve to existing files under `<wiki_dir>/` or `<raw_dir>/`.
-   - Fix or flag broken paths.
-
-2. **Recompute `mentioned_in`:**
-   - Recompute `mentioned_in` for every page under `<wiki_dir>/` by scanning all other pages' relation fields and inline links.
-   - Update the field in frontmatter.
-
-3. **Find Orphaned Pages:**
-   - Detect pages under `<wiki_dir>/` not reachable from `<wiki_dir>/index.md` or any `mentioned_in` field.
-   - Add them to the index in their appropriate category.
-
-4. **Detect Duplicate or Overlapping Pages:**
-   - Detect duplicate or heavily overlapping pages.
-   - Consolidate and update relations across the wiki graph.
-
-5. **Check Factual Inconsistencies:**
-   - Check for factual inconsistencies between pages in `contradicts` relations or covering the same topic.
-   - Clarify or resolve contradictions.
-
-6. **Identify Knowledge Gaps:**
-   - Identify knowledge gaps: concepts referenced in relation fields or prose but lacking their own atomic page.
-   - Create stub pages or flag for future ingest.
-
-7. **Suggest Source Candidates:**
-   - Suggest source candidates from `<raw_dir>/` or external searches to fill identified knowledge gaps.
-
-8. **Rebuild Navigation Index (Annotated Links):**
-   - Rebuild `<wiki_dir>/index.md` (default: `.rnex/wiki/index.md`) to reflect the current set of pages and categories accurately.
-   - Ensure every page entry maintains a concise 1-line descriptive annotation.
-
-9. **Update Hot Cache (Bounded ~500-Word Ceiling):**
-   - Refresh `<wiki_dir>/hot.md` (default: `.rnex/wiki/hot.md`), verifying the 3 structured sections (Architecture & Repo Map, Current Focus, Key Contracts).
-   - Ensure the total length remains strictly under ~500 words by pruning superseded entries.
-
-10. **Append Operation Log:**
-    - Append a lint summary to `<wiki_dir>/_log.md` (default: `.rnex/wiki/_log.md`) detailing:
-      - Broken links fixed or flagged
-      - `mentioned_in` updates
-      - Orphans and duplicates resolved
-      - Gaps identified or stubbed

package/toolkit/prompts/rnex-ingest.md

@@ -1,14 +0,0 @@
-# Operational Task: Ingest New Material
-
-You are tasked with executing the ingestion and atomic break-down workflow.
-
-## Context & Inputs
-- Source files location: `.rnex/wiki/raw/` (inclusive of active member repo `README_*.md` files)
-- Goal: Deconstruct files into individual interlinked pages in `.rnex/wiki/`.
-
-## Workflow & Guidelines Router
-1. Read the strict ingest steps and verification metrics only from `.rnex/plugins/karpathy-llm/workflows/wiki-ingest.md`.
-2. Follow styling, structural, and behavioral constraints inside `.rnex/plugins/karpathy-llm/rules/behavioral.md`.
-3. Adhere to global instruction patterns in `.rnex/ai-instructions.md`.
-
-Transform unmodified raw documents on ingestion following exactly the referenced workflow. Do not use inline logic or generate outside explicit instructions. Run 'wiki-lint' upon completion to review accuracy. Update Operation Log (`.rnex/wiki/_log.md`).

package/toolkit/prompts/rnex-lint.md

@@ -1,15 +0,0 @@
-# Operational Task: Workspace Wiki Lint & Audit
-
-You are tasked with executing the Workspace Wiki Lint workflow.
-
-## Context & Inputs
-- Target Scope: `.rnex/wiki/` (including `.rnex/wiki/raw/` and synchronized member READMEs)
-- Custom Rules & Workflows loaded: `.rnex/plugins/karpathy-llm/`
-- Current Workspace Manifest: `rnex.yaml` and `.local.rnex.yaml`
-
-## Verification Rule Router
-1. Locate and parse your validation instructions exclusively from `.rnex/plugins/karpathy-llm/workflows/wiki-lint.md`.
-2. Evaluate adherence based on the agent behavior standards described in `.rnex/plugins/karpathy-llm/rules/behavioral.md` and compiled rules in `.rnex/ai-instructions.md`.
-3. Do NOT invent new lint checks or execute uncompiled heuristic instructions.
-
-Execute the 10-step protocol defined strictly within the referenced workflow. Audit `raw_dir` items, link paths, index categorization, and relational YAML frontmatter. Provide your structured audit and write feedback output to the Audit trail in `.rnex/wiki/_log.md`.