feat(init,rgai): auto-install grepai via init, delegate rtk rgai to grepai with silent fallback

.claude/hooks/rtk-rewrite.sh

@@ -2,6 +2,7 @@
 # RTK auto-rewrite hook for Claude Code PreToolUse:Bash
 # Transparently rewrites raw commands to their rtk equivalents.
 # Outputs JSON with updatedInput to modify the command before execution.
+# Source of truth: hooks/rtk-rewrite.sh (keep .claude/hooks copy in sync)
 
 # Guards: skip silently if dependencies missing
 if ! command -v rtk &>/dev/null || ! command -v jq &>/dev/null; then
@@ -84,16 +85,24 @@ elif echo "$MATCH_CMD" | grep -qE '^cargo[[:space:]]+check([[:space:]]|$)'; then
   REWRITTEN="${ENV_PREFIX}$(echo "$CMD_BODY" | sed 's/^cargo check/rtk cargo check/')"
 elif echo "$MATCH_CMD" | grep -qE '^cargo[[:space:]]+install([[:space:]]|$)'; then
   REWRITTEN="${ENV_PREFIX}$(echo "$CMD_BODY" | sed 's/^cargo install/rtk cargo install/')"
-elif echo "$MATCH_CMD" | grep -qE '^cargo[[:space:]]+nextest([[:space:]]|$)'; then
-  REWRITTEN="${ENV_PREFIX}$(echo "$CMD_BODY" | sed 's/^cargo nextest/rtk cargo nextest/')"
 elif echo "$MATCH_CMD" | grep -qE '^cargo[[:space:]]+fmt([[:space:]]|$)'; then
   REWRITTEN="${ENV_PREFIX}$(echo "$CMD_BODY" | sed 's/^cargo fmt/rtk cargo fmt/')"
 
 # --- File operations ---
+# Search priority (mandatory): rgai > rg > grep
+# Tier 1: semantic intent search (grepai/rgai) -> rtk rgai
+# Tier 2: exact search via ripgrep -> rtk grep (rtk grep runs rg -> grep fallback internally)
+# Tier 3: exact search via grep    -> rtk grep
+elif echo "$MATCH_CMD" | grep -qE '^(grepai|rgai)[[:space:]]+search([[:space:]]|$)'; then
+  REWRITTEN="${ENV_PREFIX}$(echo "$CMD_BODY" | sed -E 's/^(grepai|rgai)[[:space:]]+search[[:space:]]+/rtk rgai /')"
+elif echo "$MATCH_CMD" | grep -qE '^rgai[[:space:]]+'; then
+  REWRITTEN="${ENV_PREFIX}$(echo "$CMD_BODY" | sed -E 's/^rgai[[:space:]]+/rtk rgai /')"
 elif echo "$MATCH_CMD" | grep -qE '^cat[[:space:]]+'; then
   REWRITTEN="${ENV_PREFIX}$(echo "$CMD_BODY" | sed 's/^cat /rtk read /')"
-elif echo "$MATCH_CMD" | grep -qE '^(rg|grep)[[:space:]]+'; then
-  REWRITTEN="${ENV_PREFIX}$(echo "$CMD_BODY" | sed -E 's/^(rg|grep) /rtk grep /')"
+elif echo "$MATCH_CMD" | grep -qE '^rg[[:space:]]+'; then
+  REWRITTEN="${ENV_PREFIX}$(echo "$CMD_BODY" | sed -E 's/^rg /rtk grep /')"
+elif echo "$MATCH_CMD" | grep -qE '^grep[[:space:]]+'; then
+  REWRITTEN="${ENV_PREFIX}$(echo "$CMD_BODY" | sed -E 's/^grep /rtk grep /')"
 elif echo "$MATCH_CMD" | grep -qE '^ls([[:space:]]|$)'; then
   REWRITTEN="${ENV_PREFIX}$(echo "$CMD_BODY" | sed 's/^ls/rtk ls/')"
 elif echo "$MATCH_CMD" | grep -qE '^tree([[:space:]]|$)'; then

.gitignore

@@ -35,4 +35,10 @@ benchmark-report.md
 *.sqlite
 *.sqlite3
 rtk_tracking.db
-claudedocs
+claudedocs
+
+# grepai local index (project-specific, not portable)
+.grepai/
+
+# Claude Code local settings (user-specific)
+.claude/settings.local.json

ARCHITECTURE.md

@@ -176,7 +176,7 @@ Database: ~/.local/share/rtk/history.db
 
 ## Module Organization
 
-### Complete Module Map (30 Modules)
+### Complete Module Map (49 Modules)
 
 ```
 ┌────────────────────────────────────────────────────────────────────────┐
@@ -191,11 +191,14 @@ GIT               git.rs            status, diff, log      85-99%     ✓
                                     branch, checkout
 
 CODE SEARCH       grep_cmd.rs       grep                   60-80%     ✓
+                  rgai_cmd.rs       rgai (semantic)        85%        ✓
+                  grepai.rs         (grepai delegation)    N/A        ✓
                   diff_cmd.rs       diff                   70-85%     ✓
                   find_cmd.rs       find                   50-70%     ✓
 
 FILE OPS          ls.rs             ls                     50-70%     ✓
                   read.rs           read                   40-90%     ✓
+                  tree.rs           (internal for ls)      N/A        ✓
 
 EXECUTION         runner.rs         err, test              60-99%     ✓
                   summary.rs        smart (heuristic)      50-80%     ✓
@@ -204,51 +207,65 @@ EXECUTION         runner.rs         err, test              60-99%     ✓
 LOGS/DATA         log_cmd.rs        log                    70-90%     ✓
                   json_cmd.rs       json                   80-95%     ✓
 
+BUILD             cargo_cmd.rs      cargo build/test       80-90%     ✓
+
 JS/TS STACK       lint_cmd.rs       lint                   84%        ✓
                   tsc_cmd.rs        tsc                    83%        ✓
                   next_cmd.rs       next                   87%        ✓
                   prettier_cmd.rs   prettier               70%        ✓
+                  format_cmd.rs     format (black/ruff)    70%        ✓
                   playwright_cmd.rs playwright             94%        ✓
                   prisma_cmd.rs     prisma                 88%        ✓
                   vitest_cmd.rs     vitest                 99.5%      ✓
                   pnpm_cmd.rs       pnpm                   70-90%     ✓
-
-CONTAINERS        container.rs      podman, docker         60-80%     ✓
-
-VCS               gh_cmd.rs         gh                     26-87%     ✓
+                  npm_cmd.rs        npm run, npx           70-80%     ✓
 
 PYTHON            ruff_cmd.rs       ruff check/format      80%+       ✓
                   pytest_cmd.rs     pytest                 90%+       ✓
-                  pip_cmd.rs        pip list/outdated      70-85%     ✓
+                  pip_cmd.rs        pip/uv                 70-85%     ✓
 
-GO                go_cmd.rs         go test/build/vet      75-90%     ✓
+GO                go_cmd.rs         go test/build/vet      80-90%     ✓
                   golangci_cmd.rs   golangci-lint          85%        ✓
 
-NETWORK           wget_cmd.rs       wget                   85-95%     ✓
+CONTAINERS        container.rs      podman, docker         60-80%     ✓
+
+VCS               gh_cmd.rs         gh                     26-87%     ✓
+
+NETWORK           curl_cmd.rs       curl                   70%        ✓
+                  wget_cmd.rs       wget                   85-95%     ✓
 
 DEPENDENCIES      deps.rs           deps                   80-90%     ✓
 
 ENVIRONMENT       env_cmd.rs        env                    60-80%     ✓
 
+ANALYTICS         cc_economics.rs   cc-economics           N/A        ✓
+                  ccusage.rs        (internal)             N/A        ✓
+                  display_helpers.rs (internal)            N/A        ✓
+                  discover/         discover               N/A        ✓
+                  learn/            (error learning)       N/A        ✓
+
 SYSTEM            init.rs           init                   N/A        ✓
                   gain.rs           gain                   N/A        ✓
                   config.rs         (internal)             N/A        ✓
 
 SHARED            utils.rs          Helpers                N/A        ✓
                   filter.rs         Language filters       N/A        ✓
                   tracking.rs       Token tracking         N/A        ✓
+                  parser/           Output parser fwk      N/A        ✓
+                  rgai_arg_tests    (test-only)            N/A        ✓
 ```
 
-**Total: 46 modules** (29 command modules + 17 infrastructure modules)
+**Total: 49 modules** (33 command modules + 16 infrastructure/internal modules)
 
 ### Module Count Breakdown
 
-- **Command Modules**: 29 (directly exposed to users)
-- **Infrastructure Modules**: 17 (utils, filter, tracking, config, init, gain, etc.)
+- **Command Modules**: 33 (directly exposed to users)
+- **Infrastructure Modules**: 16 (utils, filter, tracking, config, init, gain, parser, display, etc.)
 - **Git Commands**: 7 operations (status, diff, log, add, commit, push, branch/checkout)
-- **JS/TS Tooling**: 8 modules (modern frontend/fullstack development)
-- **Python Tooling**: 3 modules (ruff, pytest, pip)
-- **Go Tooling**: 2 modules (go test/build/vet, golangci-lint)
+- **JS/TS Tooling**: 10 modules (modern frontend/fullstack development)
+- **Python Tooling**: 3 modules (ruff, pytest, pip/uv)
+- **Go Tooling**: 2 modules (go, golangci-lint)
+- **Semantic Search**: 2 modules (rgai_cmd + grepai delegation)
 
 ---
 

INSTALL.md

@@ -69,6 +69,15 @@ rtk gain  # MUST show token savings, not "command not found"
 
 ## Project Initialization
 
+### Search Priority
+
+**Search priority: rgai > rg > grep.**
+
+- Use `rtk rgai` first for semantic/intention-based discovery.
+- Use `rtk grep` for exact/regex matching.
+- `rtk grep` internally follows `rg -> grep` backend fallback.
+- For fully raw output, use `rtk proxy <cmd>`.
+
 ### Recommended: Global Hook-First Setup
 
 **Best for: All projects, automatic RTK usage**
@@ -229,7 +238,8 @@ cp ~/.claude/settings.json.bak ~/.claude/settings.json
 ```bash
 rtk ls .              # Compact tree view
 rtk read file.rs      # Optimized reading
-rtk grep "pattern" .  # Grouped search results
+rtk rgai "query"      # Semantic code search
+rtk grep "pattern" .  # Exact/regex search (internal rg -> grep fallback)
 ```
 
 ### Git
@@ -332,6 +342,7 @@ Before each session:
 - [ ] Verify RTK is installed: `rtk --version`
 - [ ] If not installed → follow "Install from fork"
 - [ ] If project not initialized → `rtk init`
+- [ ] Search: use `rtk rgai` first, then `rtk grep` for exact/regex
 - [ ] Use `rtk` for ALL git/pnpm/test/vitest commands
 - [ ] Check savings: `rtk gain`
 

README.md

@@ -51,6 +51,15 @@ With rtk: **~45,000 tokens** → **70% reduction**
 
 > Estimates based on medium-sized TypeScript/Rust projects. Actual savings vary by project size.
 
+## Search Priority
+
+**Search priority: rgai > rg > grep.**
+
+- Use `rtk rgai` first for semantic/intention-based discovery.
+- Use `rtk grep` for exact/regex matching.
+- `rtk grep` internally follows `rg -> grep` backend fallback.
+- For fully raw output, use `rtk proxy <cmd>`.
+
 ## Installation
 
 ### ⚠️ Pre-Installation Check (REQUIRED)
@@ -133,7 +142,8 @@ rtk read file.rs                # Smart file reading
 rtk read file.rs -l aggressive  # Signatures only (strips bodies)
 rtk smart file.rs               # 2-line heuristic code summary
 rtk find "*.rs" .               # Compact find results
-rtk grep "pattern" .            # Grouped search results
+rtk rgai "auth token refresh"   # Semantic code search
+rtk grep "pattern" .            # Exact/regex search (internal rg -> grep fallback)
 ```
 
 ### Git
@@ -425,6 +435,11 @@ The most effective way to use rtk is with the **auto-rewrite hook** for Claude C
 
 **Result**: 100% rtk adoption across all conversations and subagents, zero token overhead in Claude's context.
 
+**Search ladder**:
+- `rtk rgai` for semantic discovery
+- `rtk grep` for exact/regex follow-up (`rg -> grep` fallback)
+- `rtk proxy ...` when you need fully raw behavior
+
 ### What Are Hooks?
 
 **For Beginners**:
@@ -516,12 +531,16 @@ The hook is included in this repository at `.claude/hooks/rtk-rewrite.sh`. To us
 
 ### Commands Rewritten
 
+Search rewrite order is strict and deterministic: `rgai > rg > grep`.
+
 | Raw Command | Rewritten To |
 |-------------|-------------|
 | `git status/diff/log/add/commit/push/pull/branch/fetch/stash` | `rtk git ...` |
 | `gh pr/issue/run` | `rtk gh ...` |
 | `cargo test/build/clippy` | `rtk cargo ...` |
 | `cat <file>` | `rtk read <file>` |
+| `grepai/rgai search <query>` | `rtk rgai <query>` |
+| `rgai <query>` | `rtk rgai <query>` |
 | `rg/grep <pattern>` | `rtk grep <pattern>` |
 | `ls` | `rtk ls` |
 | `vitest/pnpm test` | `rtk vitest run` |