agent-wiki-cli 0.3.41__tar.gz → 0.5.0__tar.gz

{agent_wiki_cli-0.3.41/src/agent_wiki_cli.egg-info → agent_wiki_cli-0.5.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agent-wiki-cli
-Version: 0.3.41
+Version: 0.5.0
 Summary: CLI tool to maintain hybrid LLM Wikis for multi-language projects.
 Author-email: Denis Sivagin <denissvgn@gmail.com>
 License: MIT
@@ -58,15 +58,16 @@ external tools when they are available on `PATH`.
 | Area | Implementation | Runtime requirement |
 |---|---|---|
 | Python | stdlib `ast` | Python 3.9+ |
-| TypeScript / TSX | bundled Node script using `ts-morph` | Node.js and npm |
-| Go | bundled Go extractor using `go/ast` | Go toolchain |
-| Rust | bundled Rust extractor using `syn` | Cargo / Rust toolchain |
+| TypeScript / TSX | bundled Node script using `ts-morph` | prepared Node.js dependencies |
+| Go | bundled Go extractor using `go/ast` | prepared helper binary |
+| Rust | bundled Rust extractor using `syn` | prepared helper binary |
 | Docker / Compose | built-in parsers | none |
 | MCP server | official Python MCP SDK | `agent-wiki-cli[mcp]`, Python 3.10+ |
 
-TypeScript, Go, and Rust extras are metadata-only; the actual toolchains must be
-installed separately. The TypeScript extractor runs `npm install` in its bundled
-extractor directory on first use if `node_modules` is missing.
+TypeScript, Go, and Rust extras are metadata-only; prepare their helper
+dependencies explicitly with `llm-wiki prepare-extractors`. Lint, CI, and
+extract never run `npm install`, `go build`, `go run`, `cargo build`, or
+`cargo run` automatically.
 
 ## Agent Support
 
@@ -232,16 +233,23 @@ manifest.
 
 ```bash
 llm-wiki sync --src-dir . --wiki-dir docs/llm_wiki
+llm-wiki sync --jobs auto --cache-stats --src-dir . --wiki-dir docs/llm_wiki
 ```
 
 If an older wiki has `index.md` but no manifest, `sync` seeds
 `.llm-wiki-manifest.json` without modifying pages. If neither a manifest nor an
-existing wiki is present, run `bootstrap` first.
+existing wiki is present, run `bootstrap` first. Sync uses the same safe
+persistent inventory cache as lint when a git directory is available. Use
+`--no-cache`, `--rebuild-cache`, `--cache-dir PATH`, and `--cache-stats` to
+control or inspect cache behavior. Use `--jobs N` or `--jobs auto` to opt into
+parallel built-in language extraction. Sync repairs manifests with invalid
+source hashes without touching pages, and stops unusually broad diffs unless
+`--force` is used.
 
 ### `extract`
 
 Print source inventory as JSON. All registered extractors run; missing optional
-toolchains are skipped when there are no matching source files.
+prepared helpers are skipped when there are no matching source files.
 
 ```bash
 llm-wiki extract --src-dir .
@@ -253,6 +261,23 @@ llm-wiki extract --src-dir . --package llm_wiki_cli
 llm-wiki extract --src-dir . --include-empty
 ```
 
+### `prepare-extractors`
+
+Prepare TypeScript dependencies and cached Go/Rust helper binaries outside the
+lint/extract hot path.
+
+```bash
+llm-wiki prepare-extractors --src-dir .
+llm-wiki prepare-extractors --language typescript --language go
+llm-wiki prepare-extractors --cache-dir .cache/llm-wiki
+```
+
+When `--language` is omitted, only helper languages detected in `--src-dir` are
+prepared. Helper cache resolution follows `--cache-dir`, then
+`LLM_WIKI_CACHE_DIR`, then `.git/llm-wiki-extractors/`. If Go is installed in a
+nonstandard location or the `go` on `PATH` cannot run, set
+`LLM_WIKI_GO=/path/to/go` before running `prepare-extractors`.
+
 ### `lint` and `ci-check`
 
 Validate wiki links, orphan pages, entities, modules, workflows,
@@ -261,20 +286,40 @@ infrastructure, plugin lint rules, and team policy.
 ```bash
 llm-wiki lint --wiki-dir docs/llm_wiki --src-dir .
 llm-wiki lint --strict --wiki-dir docs/llm_wiki --src-dir .
+llm-wiki lint --profile --wiki-dir docs/llm_wiki --src-dir .
+llm-wiki lint --cache-stats --wiki-dir docs/llm_wiki --src-dir .
+llm-wiki lint --jobs auto --wiki-dir docs/llm_wiki --src-dir .
 ```
 
 Strict mode also requires the core wiki structure and a fresh sync manifest.
+`--profile` suppresses the human-readable lint text and prints one JSON object
+to stdout containing the normal lint report, diagnostics, and phase timings.
+The JSON contract is preserved for extractor failures as well; lint still exits
+nonzero, but stdout remains machine-readable.
+Lint uses a persistent deep-inventory cache by default when a git directory is
+available, storing `.git/llm-wiki-inventory-cache.json`. Override the cache
+directory with `LLM_WIKI_CACHE_DIR` or `--cache-dir PATH`; the CLI flag wins.
+Use `--no-cache` to disable load/save, `--rebuild-cache` to ignore and rewrite
+the cache, and `--cache-stats` to include cache diagnostics. Cache corruption or
+invalid fingerprints fall back to a full extraction without reducing lint
+coverage. With `--profile --cache-stats`, the JSON payload includes a top-level
+`cache` object. Use `--jobs N` or `--jobs auto` to opt into parallel built-in
+language extraction; the default is `--jobs 1`. Plugin extractors remain
+sequential unless future plugin metadata explicitly marks them parallel-safe.
 
 For CI:
 
 ```bash
 llm-wiki ci-check --src-dir . --wiki-dir docs/llm_wiki
+llm-wiki ci-check --jobs auto --src-dir . --wiki-dir docs/llm_wiki
 llm-wiki ci-check --format json --report .git/llm-wiki-ci-report.md
 llm-wiki ci-check --format markdown
 ```
 
 `ci-check` always runs strict validation, writes a Markdown report, records a
-local metrics event, and exits nonzero on validation failure.
+local metrics event, uses the same safe inventory cache when available, and
+exits nonzero on validation failure. `--report` is an output path, so explicit
+absolute paths and relative artifact paths outside the project root are allowed.
 
 ### `context`
 

{agent_wiki_cli-0.3.41 → agent_wiki_cli-0.5.0}/README.md

@@ -28,15 +28,16 @@ external tools when they are available on `PATH`.
 | Area | Implementation | Runtime requirement |
 |---|---|---|
 | Python | stdlib `ast` | Python 3.9+ |
-| TypeScript / TSX | bundled Node script using `ts-morph` | Node.js and npm |
-| Go | bundled Go extractor using `go/ast` | Go toolchain |
-| Rust | bundled Rust extractor using `syn` | Cargo / Rust toolchain |
+| TypeScript / TSX | bundled Node script using `ts-morph` | prepared Node.js dependencies |
+| Go | bundled Go extractor using `go/ast` | prepared helper binary |
+| Rust | bundled Rust extractor using `syn` | prepared helper binary |
 | Docker / Compose | built-in parsers | none |
 | MCP server | official Python MCP SDK | `agent-wiki-cli[mcp]`, Python 3.10+ |
 
-TypeScript, Go, and Rust extras are metadata-only; the actual toolchains must be
-installed separately. The TypeScript extractor runs `npm install` in its bundled
-extractor directory on first use if `node_modules` is missing.
+TypeScript, Go, and Rust extras are metadata-only; prepare their helper
+dependencies explicitly with `llm-wiki prepare-extractors`. Lint, CI, and
+extract never run `npm install`, `go build`, `go run`, `cargo build`, or
+`cargo run` automatically.
 
 ## Agent Support
 
@@ -202,16 +203,23 @@ manifest.
 
 ```bash
 llm-wiki sync --src-dir . --wiki-dir docs/llm_wiki
+llm-wiki sync --jobs auto --cache-stats --src-dir . --wiki-dir docs/llm_wiki
 ```
 
 If an older wiki has `index.md` but no manifest, `sync` seeds
 `.llm-wiki-manifest.json` without modifying pages. If neither a manifest nor an
-existing wiki is present, run `bootstrap` first.
+existing wiki is present, run `bootstrap` first. Sync uses the same safe
+persistent inventory cache as lint when a git directory is available. Use
+`--no-cache`, `--rebuild-cache`, `--cache-dir PATH`, and `--cache-stats` to
+control or inspect cache behavior. Use `--jobs N` or `--jobs auto` to opt into
+parallel built-in language extraction. Sync repairs manifests with invalid
+source hashes without touching pages, and stops unusually broad diffs unless
+`--force` is used.
 
 ### `extract`
 
 Print source inventory as JSON. All registered extractors run; missing optional
-toolchains are skipped when there are no matching source files.
+prepared helpers are skipped when there are no matching source files.
 
 ```bash
 llm-wiki extract --src-dir .
@@ -223,6 +231,23 @@ llm-wiki extract --src-dir . --package llm_wiki_cli
 llm-wiki extract --src-dir . --include-empty
 ```
 
+### `prepare-extractors`
+
+Prepare TypeScript dependencies and cached Go/Rust helper binaries outside the
+lint/extract hot path.
+
+```bash
+llm-wiki prepare-extractors --src-dir .
+llm-wiki prepare-extractors --language typescript --language go
+llm-wiki prepare-extractors --cache-dir .cache/llm-wiki
+```
+
+When `--language` is omitted, only helper languages detected in `--src-dir` are
+prepared. Helper cache resolution follows `--cache-dir`, then
+`LLM_WIKI_CACHE_DIR`, then `.git/llm-wiki-extractors/`. If Go is installed in a
+nonstandard location or the `go` on `PATH` cannot run, set
+`LLM_WIKI_GO=/path/to/go` before running `prepare-extractors`.
+
 ### `lint` and `ci-check`
 
 Validate wiki links, orphan pages, entities, modules, workflows,
@@ -231,20 +256,40 @@ infrastructure, plugin lint rules, and team policy.
 ```bash
 llm-wiki lint --wiki-dir docs/llm_wiki --src-dir .
 llm-wiki lint --strict --wiki-dir docs/llm_wiki --src-dir .
+llm-wiki lint --profile --wiki-dir docs/llm_wiki --src-dir .
+llm-wiki lint --cache-stats --wiki-dir docs/llm_wiki --src-dir .
+llm-wiki lint --jobs auto --wiki-dir docs/llm_wiki --src-dir .
 ```
 
 Strict mode also requires the core wiki structure and a fresh sync manifest.
+`--profile` suppresses the human-readable lint text and prints one JSON object
+to stdout containing the normal lint report, diagnostics, and phase timings.
+The JSON contract is preserved for extractor failures as well; lint still exits
+nonzero, but stdout remains machine-readable.
+Lint uses a persistent deep-inventory cache by default when a git directory is
+available, storing `.git/llm-wiki-inventory-cache.json`. Override the cache
+directory with `LLM_WIKI_CACHE_DIR` or `--cache-dir PATH`; the CLI flag wins.
+Use `--no-cache` to disable load/save, `--rebuild-cache` to ignore and rewrite
+the cache, and `--cache-stats` to include cache diagnostics. Cache corruption or
+invalid fingerprints fall back to a full extraction without reducing lint
+coverage. With `--profile --cache-stats`, the JSON payload includes a top-level
+`cache` object. Use `--jobs N` or `--jobs auto` to opt into parallel built-in
+language extraction; the default is `--jobs 1`. Plugin extractors remain
+sequential unless future plugin metadata explicitly marks them parallel-safe.
 
 For CI:
 
 ```bash
 llm-wiki ci-check --src-dir . --wiki-dir docs/llm_wiki
+llm-wiki ci-check --jobs auto --src-dir . --wiki-dir docs/llm_wiki
 llm-wiki ci-check --format json --report .git/llm-wiki-ci-report.md
 llm-wiki ci-check --format markdown
 ```
 
 `ci-check` always runs strict validation, writes a Markdown report, records a
-local metrics event, and exits nonzero on validation failure.
+local metrics event, uses the same safe inventory cache when available, and
+exits nonzero on validation failure. `--report` is an output path, so explicit
+absolute paths and relative artifact paths outside the project root are allowed.
 
 ### `context`
 

{agent_wiki_cli-0.3.41 → agent_wiki_cli-0.5.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "agent-wiki-cli"
-version = "0.3.41"
+version = "0.5.0"
 description = "CLI tool to maintain hybrid LLM Wikis for multi-language projects."
 readme = "README.md"
 license = {text = "MIT"}

{agent_wiki_cli-0.3.41 → agent_wiki_cli-0.5.0/src/agent_wiki_cli.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agent-wiki-cli
-Version: 0.3.41
+Version: 0.5.0
 Summary: CLI tool to maintain hybrid LLM Wikis for multi-language projects.
 Author-email: Denis Sivagin <denissvgn@gmail.com>
 License: MIT
@@ -58,15 +58,16 @@ external tools when they are available on `PATH`.
 | Area | Implementation | Runtime requirement |
 |---|---|---|
 | Python | stdlib `ast` | Python 3.9+ |
-| TypeScript / TSX | bundled Node script using `ts-morph` | Node.js and npm |
-| Go | bundled Go extractor using `go/ast` | Go toolchain |
-| Rust | bundled Rust extractor using `syn` | Cargo / Rust toolchain |
+| TypeScript / TSX | bundled Node script using `ts-morph` | prepared Node.js dependencies |
+| Go | bundled Go extractor using `go/ast` | prepared helper binary |
+| Rust | bundled Rust extractor using `syn` | prepared helper binary |
 | Docker / Compose | built-in parsers | none |
 | MCP server | official Python MCP SDK | `agent-wiki-cli[mcp]`, Python 3.10+ |
 
-TypeScript, Go, and Rust extras are metadata-only; the actual toolchains must be
-installed separately. The TypeScript extractor runs `npm install` in its bundled
-extractor directory on first use if `node_modules` is missing.
+TypeScript, Go, and Rust extras are metadata-only; prepare their helper
+dependencies explicitly with `llm-wiki prepare-extractors`. Lint, CI, and
+extract never run `npm install`, `go build`, `go run`, `cargo build`, or
+`cargo run` automatically.
 
 ## Agent Support
 
@@ -232,16 +233,23 @@ manifest.
 
 ```bash
 llm-wiki sync --src-dir . --wiki-dir docs/llm_wiki
+llm-wiki sync --jobs auto --cache-stats --src-dir . --wiki-dir docs/llm_wiki
 ```
 
 If an older wiki has `index.md` but no manifest, `sync` seeds
 `.llm-wiki-manifest.json` without modifying pages. If neither a manifest nor an
-existing wiki is present, run `bootstrap` first.
+existing wiki is present, run `bootstrap` first. Sync uses the same safe
+persistent inventory cache as lint when a git directory is available. Use
+`--no-cache`, `--rebuild-cache`, `--cache-dir PATH`, and `--cache-stats` to
+control or inspect cache behavior. Use `--jobs N` or `--jobs auto` to opt into
+parallel built-in language extraction. Sync repairs manifests with invalid
+source hashes without touching pages, and stops unusually broad diffs unless
+`--force` is used.
 
 ### `extract`
 
 Print source inventory as JSON. All registered extractors run; missing optional
-toolchains are skipped when there are no matching source files.
+prepared helpers are skipped when there are no matching source files.
 
 ```bash
 llm-wiki extract --src-dir .
@@ -253,6 +261,23 @@ llm-wiki extract --src-dir . --package llm_wiki_cli
 llm-wiki extract --src-dir . --include-empty
 ```
 
+### `prepare-extractors`
+
+Prepare TypeScript dependencies and cached Go/Rust helper binaries outside the
+lint/extract hot path.
+
+```bash
+llm-wiki prepare-extractors --src-dir .
+llm-wiki prepare-extractors --language typescript --language go
+llm-wiki prepare-extractors --cache-dir .cache/llm-wiki
+```
+
+When `--language` is omitted, only helper languages detected in `--src-dir` are
+prepared. Helper cache resolution follows `--cache-dir`, then
+`LLM_WIKI_CACHE_DIR`, then `.git/llm-wiki-extractors/`. If Go is installed in a
+nonstandard location or the `go` on `PATH` cannot run, set
+`LLM_WIKI_GO=/path/to/go` before running `prepare-extractors`.
+
 ### `lint` and `ci-check`
 
 Validate wiki links, orphan pages, entities, modules, workflows,
@@ -261,20 +286,40 @@ infrastructure, plugin lint rules, and team policy.
 ```bash
 llm-wiki lint --wiki-dir docs/llm_wiki --src-dir .
 llm-wiki lint --strict --wiki-dir docs/llm_wiki --src-dir .
+llm-wiki lint --profile --wiki-dir docs/llm_wiki --src-dir .
+llm-wiki lint --cache-stats --wiki-dir docs/llm_wiki --src-dir .
+llm-wiki lint --jobs auto --wiki-dir docs/llm_wiki --src-dir .
 ```
 
 Strict mode also requires the core wiki structure and a fresh sync manifest.
+`--profile` suppresses the human-readable lint text and prints one JSON object
+to stdout containing the normal lint report, diagnostics, and phase timings.
+The JSON contract is preserved for extractor failures as well; lint still exits
+nonzero, but stdout remains machine-readable.
+Lint uses a persistent deep-inventory cache by default when a git directory is
+available, storing `.git/llm-wiki-inventory-cache.json`. Override the cache
+directory with `LLM_WIKI_CACHE_DIR` or `--cache-dir PATH`; the CLI flag wins.
+Use `--no-cache` to disable load/save, `--rebuild-cache` to ignore and rewrite
+the cache, and `--cache-stats` to include cache diagnostics. Cache corruption or
+invalid fingerprints fall back to a full extraction without reducing lint
+coverage. With `--profile --cache-stats`, the JSON payload includes a top-level
+`cache` object. Use `--jobs N` or `--jobs auto` to opt into parallel built-in
+language extraction; the default is `--jobs 1`. Plugin extractors remain
+sequential unless future plugin metadata explicitly marks them parallel-safe.
 
 For CI:
 
 ```bash
 llm-wiki ci-check --src-dir . --wiki-dir docs/llm_wiki
+llm-wiki ci-check --jobs auto --src-dir . --wiki-dir docs/llm_wiki
 llm-wiki ci-check --format json --report .git/llm-wiki-ci-report.md
 llm-wiki ci-check --format markdown
 ```
 
 `ci-check` always runs strict validation, writes a Markdown report, records a
-local metrics event, and exits nonzero on validation failure.
+local metrics event, uses the same safe inventory cache when available, and
+exits nonzero on validation failure. `--report` is an output path, so explicit
+absolute paths and relative artifact paths outside the project root are allowed.
 
 ### `context`
 

{agent_wiki_cli-0.3.41 → agent_wiki_cli-0.5.0}/src/agent_wiki_cli.egg-info/SOURCES.txt

@@ -26,6 +26,7 @@ src/llm_wiki_cli/commands/metrics_cmd.py
 src/llm_wiki_cli/commands/migrate_cmd.py
 src/llm_wiki_cli/commands/obsidian_cmd.py
 src/llm_wiki_cli/commands/plugins_cmd.py
+src/llm_wiki_cli/commands/prepare_extractors_cmd.py
 src/llm_wiki_cli/commands/release_cmd.py
 src/llm_wiki_cli/commands/review_cmd.py
 src/llm_wiki_cli/commands/status_cmd.py
@@ -49,7 +50,9 @@ src/llm_wiki_cli/extractors/ts_scripts/extract.js
 src/llm_wiki_cli/extractors/ts_scripts/package.json
 src/llm_wiki_cli/services/__init__.py
 src/llm_wiki_cli/services/circuit_breaker.py
+src/llm_wiki_cli/services/extractor_helpers.py
 src/llm_wiki_cli/services/imports.py
+src/llm_wiki_cli/services/inventory_cache.py
 src/llm_wiki_cli/services/io.py
 src/llm_wiki_cli/services/lockfile.py
 src/llm_wiki_cli/services/mcp_server.py
@@ -60,10 +63,12 @@ src/llm_wiki_cli/services/paths.py
 src/llm_wiki_cli/services/plugins.py
 src/llm_wiki_cli/services/schema.py
 src/llm_wiki_cli/services/secure_file.py
+src/llm_wiki_cli/services/source_snapshot.py
 src/llm_wiki_cli/services/team.py
 src/llm_wiki_cli/services/versioning.py
 tests/test_bootstrap.py
 tests/test_bump.py
+tests/test_ci_check.py
 tests/test_circuit_breaker.py
 tests/test_config.py
 tests/test_context.py
@@ -72,10 +77,13 @@ tests/test_docker_extract.py
 tests/test_docker_lint.py
 tests/test_e2e.py
 tests/test_extract.py
+tests/test_extractor_helpers.py
 tests/test_generate_prompt.py
 tests/test_go_extract.py
 tests/test_hook.py
+tests/test_imports.py
 tests/test_init.py
+tests/test_inventory_cache.py
 tests/test_io.py
 tests/test_lint.py
 tests/test_lockfile.py
@@ -89,6 +97,7 @@ tests/test_plugins.py
 tests/test_release.py
 tests/test_rust_extract.py
 tests/test_schema.py
+tests/test_source_snapshot.py
 tests/test_status.py
 tests/test_sync.py
 tests/test_team.py

{agent_wiki_cli-0.3.41 → agent_wiki_cli-0.5.0}/src/llm_wiki_cli/cli.py

@@ -17,6 +17,7 @@ from .commands import (
     migrate_cmd,
     obsidian_cmd,
     plugins_cmd,
+    prepare_extractors_cmd,
     release_cmd,
     review_cmd,
     status_cmd,
@@ -37,6 +38,18 @@ def _positive_int(value: str) -> int:
     return parsed
 
 
+def _jobs_value(value: str) -> int:
+    if value == "auto":
+        return max(1, os.cpu_count() or 1)
+    try:
+        parsed = int(value)
+    except ValueError as exc:
+        raise argparse.ArgumentTypeError("must be a positive integer or 'auto'") from exc
+    if parsed < 1:
+        raise argparse.ArgumentTypeError("must be greater than zero")
+    return parsed
+
+
 def main():
     parser = argparse.ArgumentParser(description="LLM Wiki CLI")
     parser.add_argument("--version", action="version", version=f"llm-wiki {__version__}")
@@ -72,6 +85,33 @@ def main():
     lint_parser.add_argument("--src-dir", default=".", help="Source directory to cross-reference against")
     lint_parser.add_argument("--strict", action="store_true",
                              help="Require core wiki structure and a fresh sync manifest")
+    lint_parser.add_argument("--profile", action="store_true",
+                             help="Print combined lint report and phase timings as JSON")
+    lint_parser.add_argument("--no-cache", action="store_true",
+                             help="Disable persistent inventory cache for this run")
+    lint_parser.add_argument("--rebuild-cache", action="store_true",
+                             help="Ignore existing inventory cache and rewrite it after extraction")
+    lint_parser.add_argument("--cache-stats", action="store_true",
+                             help="Include inventory cache diagnostics in lint output")
+    lint_parser.add_argument("--cache-dir", default=None, metavar="PATH",
+                             help="Directory for llm-wiki-inventory-cache.json")
+    lint_parser.add_argument("--jobs", type=_jobs_value, default=1, metavar="JOBS",
+                             help="Parallel built-in extractor jobs: positive integer or 'auto' (default: 1)")
+
+    # prepare-extractors command
+    prepare_parser = subparsers.add_parser(
+        "prepare-extractors",
+        help="Prepare TypeScript, Go, and Rust extractor helpers",
+    )
+    prepare_parser.add_argument("--src-dir", default=".", help="Source directory to inspect")
+    prepare_parser.add_argument("--cache-dir", default=None, metavar="PATH",
+                                help="Directory for extractor helper cache")
+    prepare_parser.add_argument(
+        "--language",
+        action="append",
+        choices=["typescript", "go", "rust"],
+        help="Helper language to prepare; may be repeated",
+    )
 
     # ci-check command
     ci_parser = subparsers.add_parser("ci-check", help="Run strict wiki validation and write a CI report")
@@ -82,6 +122,8 @@ def main():
                            help="Console output format (default: text)")
     ci_parser.add_argument("--report", default=".git/llm-wiki-ci-report.md",
                            help="Markdown report path (default: .git/llm-wiki-ci-report.md)")
+    ci_parser.add_argument("--jobs", type=_jobs_value, default=1, metavar="JOBS",
+                           help="Parallel built-in extractor jobs: positive integer or 'auto' (default: 1)")
 
     # hook command
     hook_parser = subparsers.add_parser("install-hook", help="Install git hooks for wiki sync")
@@ -306,6 +348,18 @@ def main():
     sync_parser.add_argument("--src-dir", default=".", help="Source directory to scan (default: .)")
     sync_parser.add_argument("--wiki-dir", default=DEFAULT_WIKI_DIR,
                              help="Wiki directory (default: docs/llm_wiki)")
+    sync_parser.add_argument("--no-cache", action="store_true",
+                             help="Disable persistent inventory cache for this run")
+    sync_parser.add_argument("--rebuild-cache", action="store_true",
+                             help="Ignore existing inventory cache and rewrite it after extraction")
+    sync_parser.add_argument("--cache-stats", action="store_true",
+                             help="Include inventory cache diagnostics in sync output")
+    sync_parser.add_argument("--cache-dir", default=None, metavar="PATH",
+                             help="Directory for llm-wiki-inventory-cache.json")
+    sync_parser.add_argument("--jobs", type=_jobs_value, default=1, metavar="JOBS",
+                             help="Parallel built-in extractor jobs: positive integer or 'auto' (default: 1)")
+    sync_parser.add_argument("--force", action="store_true",
+                             help="Allow sync to apply unusually broad source diffs")
 
     # migrate command
     migrate_parser = subparsers.add_parser(
@@ -349,6 +403,8 @@ def main():
             extract_cmd.run(args)
         elif args.command == "lint":
             lint_cmd.run(args)
+        elif args.command == "prepare-extractors":
+            prepare_extractors_cmd.run(args)
         elif args.command == "ci-check":
             ci_check_cmd.run(args)
         elif args.command == "install-hook":

{agent_wiki_cli-0.3.41 → agent_wiki_cli-0.5.0}/src/llm_wiki_cli/commands/bootstrap_cmd.py

@@ -11,9 +11,10 @@ from pathlib import Path
 
 from .extract_cmd import get_call_graph, get_docker_inventory, get_inventory_result, print_inventory_failures
 from ..config import validate_path
-from ..services.imports import module_path_candidates
+from ..services.imports import ModulePathResolver, build_module_path_resolver
 from ..services.io import read_md, write_md
 from ..services.paths import normalize_source_path
+from ..services.source_snapshot import build_source_snapshot
 
 
 def _module_name_from_path(filepath: str) -> str:
@@ -128,7 +129,13 @@ def build_entity_page_map(inventory: dict) -> dict[tuple[str, str], str]:
     return page_map
 
 
-def _build_relationships(inventory: dict, module_page_map: dict[str, str] | None = None) -> dict:
+def _build_relationships(
+    inventory: dict,
+    module_page_map: dict[str, str] | None = None,
+    *,
+    target_entities: set[tuple[str, str]] | None = None,
+    resolver: ModulePathResolver | None = None,
+) -> dict:
     """Cross-reference imports against known entity identities to build a usage graph.
 
     Returns a dict mapping ``(entity_name, defining_filepath)`` to a list of
@@ -148,6 +155,7 @@ def _build_relationships(inventory: dict, module_page_map: dict[str, str] | None
     # relationship map: (entity_name, defining_filepath) -> relationship records
     relationships = defaultdict(list)
     _mod_page_map = module_page_map or {}
+    module_resolver = resolver or build_module_path_resolver(inventory)
 
     for filepath, data in inventory.items():
         mod_name = _module_name_from_path(filepath)
@@ -158,19 +166,20 @@ def _build_relationships(inventory: dict, module_page_map: dict[str, str] | None
             if not entity_name or entity_name not in entity_to_files:
                 continue
             candidates = set(entity_to_files[entity_name])
-            module_candidates = module_path_candidates(
-                imp.get("module", ""), filepath, inventory
-            )
+            module_candidates = module_resolver.candidates(imp.get("module", ""), filepath)
             if module_candidates:
                 candidates &= module_candidates
             candidates.discard(filepath)
             if len(candidates) != 1:
                 continue
             defining_filepath = next(iter(candidates))
+            entity_key = (entity_name, defining_filepath)
+            if target_entities is not None and entity_key not in target_entities:
+                continue
             visible_names = {entity_name}
             if imp.get("alias"):
                 visible_names.add(imp["alias"])
-            imported_entities[(entity_name, defining_filepath)] = visible_names
+            imported_entities[entity_key] = visible_names
 
         for entity_key, visible_names in imported_entities.items():
             for fn in data.get("functions", []):
@@ -824,19 +833,26 @@ def run(args):
     deep = depth == "full"
     skip_workflows = getattr(args, "skip_workflows", False)
 
-    print(f"Bootstrapping wiki from source: {src_dir} (depth={depth})")
-    print(f"Wiki output directory: {wiki_dir}")
+    print(f"Bootstrapping wiki from source: {src_dir} (depth={depth})", flush=True)
+    print(f"Wiki output directory: {wiki_dir}", flush=True)
 
     # Ensure wiki structure exists
     for subdir in ["entities", "modules", "workflows", "infrastructure"]:
         (wiki_dir / subdir).mkdir(parents=True, exist_ok=True)
 
     # 1. Extract full AST inventory
-    inventory_result = get_inventory_result(src_dir, deep=deep)
+    print("Extracting source inventory...", flush=True)
+    source_snapshot = build_source_snapshot(src_dir)
+    inventory_result = get_inventory_result(
+        src_dir,
+        deep=deep,
+        source_snapshot=source_snapshot,
+    )
     if inventory_result.failed:
         print_inventory_failures(inventory_result)
         sys.exit(1)
     inventory = inventory_result.inventory
+    print(f"Extracted source inventory: {len(inventory)} file(s).", flush=True)
 
     if not inventory:
         print("No supported source files with classes or functions found. Nothing to bootstrap.")
@@ -855,8 +871,13 @@ def run(args):
     _entity_page_name_cache: dict = build_entity_page_map(inventory)
 
     # 2. Build cross-reference relationships (only meaningful in deep mode)
+    if deep:
+        print("Building cross-reference relationships...", flush=True)
     relationships = _build_relationships(inventory, _module_page_map) if deep else {}
+    if deep:
+        print(f"Built cross-reference relationships: {sum(len(v) for v in relationships.values())}.", flush=True)
 
+    print("Generating entity and module pages...", flush=True)
     for filepath, file_data in inventory.items():
         mod_page_name = _module_page_map[filepath]
         # Map cls_name -> page_stem for classes in this file (used by module page links)
@@ -893,6 +914,10 @@ def run(args):
             "path": filepath,
             "docstring": file_data.get("module_docstring", ""),
         })
+    print(
+        f"Generated entity/module pages: {entities_created} entities, {modules_created} modules.",
+        flush=True,
+    )
 
     # 3. Generate workflow pages from call graph (deep mode only)
     workflow_entries = []
@@ -912,7 +937,8 @@ def run(args):
     # 4. Generate infrastructure pages (Dockerfile, docker-compose, etc.)
     infra_entries = []
     infra_created = 0
-    docker_inventory = get_docker_inventory(src_dir)
+    print("Generating infrastructure pages...", flush=True)
+    docker_inventory = get_docker_inventory(src_dir, source_snapshot=source_snapshot)
     for docker_file, docker_info in docker_inventory.items():
         page_name = docker_file.replace("\\", "/").replace("/", "_").replace(".", "_")
         infra_path = wiki_dir / "infrastructure" / f"{page_name}.md"
@@ -923,6 +949,7 @@ def run(args):
             infra_created += 1
             print(f"  CREATE infrastructure: {page_name}")
         infra_entries.append({"name": page_name, "type": docker_info["type"]})
+    print(f"Generated infrastructure pages: {infra_created}.", flush=True)
 
     # 5. Rebuild index.md
     index_path = wiki_dir / "index.md"
@@ -969,6 +996,7 @@ def run(args):
     manifest = SyncManifest.build_from_inventory(
         inventory, src_dir, _entity_page_name_cache, _module_page_map,
     )
+    print("Writing sync manifest...", flush=True)
     manifest.save(wiki_dir)
     print(f"  WRITE {wiki_dir / '.llm-wiki-manifest.json'}")