Docs Commands

Commands for searching and reading B2C Commerce documentation spanning multiple corpora: Script API reference (dw.* classes/modules), Developer Center guides (conceptual/how-to content across Commerce API, PWA Kit, SFRA, and more), Salesforce Help (administration and merchandising content from help.salesforce.com), tooling documentation (this project's own guides), the standard (system) job step catalog, and bundled XSD schemas. Download fresh Script API documentation from an instance.

Use these commands for any B2C Commerce developer or administrator question that is not already grounded in your project context — from Script API method lookups to SCAPI integration patterns, job step parameters, storefront framework guides, Business Manager administration, and authentication flows.

The corpus searched by docs search / docs read includes:

Script API reference — dw.* classes/modules for server-side scripting (category: script-api). Content is read online from developer.salesforce.com (cached locally after first read); the search index is bundled.
Developer Center guides — conceptual and how-to content from B2C Commerce Developer Center, organized into categories: commerce-api, pwa-kit-managed-runtime, sfnext, sfra, b2c-commerce
Salesforce Help — administration and merchandising documentation from help.salesforce.com, split into two categories: help-admin (import/export, jobs, replication, security, Account Manager, permissions, logs, inventory ops) and help-merchant (catalogs, products, promotions, search, content, analytics, SEO). Each entry's url links the live Help article; content is read from the raw markdown source.
Tooling guides — documentation for this CLI, MCP server, SDK, and VS Code extension (category: tooling)
Standard job steps — built-in job step type IDs (for example ImportCatalog, ExportCatalog, ImportInventoryLists) that you add to Business Manager job flows (category: job-step). Read the catalog overview with b2c docs read job-steps, or a specific step with b2c docs read <TypeID>. See the b2c-cli:b2c-job and b2c:b2c-custom-job-steps skills for how standard steps fit into job flows.
XSD schemas — import/export data format definitions

Search results include category, summary, keywords, url, and sourceUrl fields to help triage matches before reading full content. Each documentation entry with a public page carries both url (the .html page for browser viewing) and sourceUrl (the raw .md source). Content for online corpora (Script API, Developer Center guides, Salesforce Help) is fetched from the sourceUrl on docs read and cached locally; if a fetch fails, the command falls back to the indexed summary + headings and prints both URLs so you can retrieve the page yourself. Standard job steps are the one corpus whose content ships bundled (they have no public page).

Authentication

Operation	Auth Required
`docs search`	None (uses the local bundled index)
`docs read`	None (bundled job steps + online fetch for Script API, guides, and Help)
`docs schema`	None (uses local bundled schemas)
`docs download`	Instance + WebDAV credentials

For b2c docs download, configure instance and WebDAV access:

bash

export SFCC_SERVER=my-sandbox.demandware.net
export SFCC_USERNAME=your-username
export SFCC_PASSWORD=your-password

In addition to these topic-specific options, all commands also support global flags.

b2c docs search

Search documentation across all corpora using content-aware BM25-style ranking. Results include metadata (category, summary, keywords, url) to help triage matches.

Usage

bash

b2c docs search [query]

Arguments

Argument	Description	Required
`query`	Search query (keywords, class name, or topic)	No (required unless `--list` is used)

Flags

Flag	Description	Default
`--limit`, `-l`	Maximum number of results to display	`20`
`--category`, `-c`	Filter by category: `script-api`, `commerce-api`, `pwa-kit-managed-runtime`, `sfnext`, `sfra`, `b2c-commerce`, `tooling`, `job-step`, `help-admin`, `help-merchant`	(none)
`--workspace`	Workspace awareness: `auto` (default, auto-detects project type), `all` (disable), or specify one or more types comma-separated: `cartridges`, `sfra`, `pwa-kit-v3`, `storefront-next`. Boosts relevant categories and de-boosts competing storefront frameworks	`auto`
`--topics`	Allowlist that bounds the whole corpus to these categories (comma-separated; env `SFCC_DOCS_TOPICS`). `--category`/`--workspace` narrow within it; unknown names are ignored with a warning	(all)
`--list`	List all available documentation entries	`false`
`--columns`	Columns to display (comma-separated). Available: id, title, category, summary, keywords, url, sourceUrl, score	`id, category, title, score`
`--extended`, `-x`	Show all columns including extended fields	`false`

Examples

bash

# Search across all documentation
b2c docs search "passwordless login"

# Filter by category for targeted results
b2c docs search "passwordless login" --category commerce-api

# Search Developer Center guides
b2c docs search "storefront components" --category sfra

# Search Script API reference
b2c docs search ProductMgr --category script-api

# Search PWA Kit / MRT documentation
b2c docs search "deploy bundle" --category pwa-kit-managed-runtime

# Search Salesforce Help administration content (import/export, jobs, security, ...)
b2c docs search "site import export" --category help-admin

# Search Salesforce Help merchandising content (catalogs, promotions, search, ...)
b2c docs search "price book assignment" --category help-merchant

# Find a standard job step by name
b2c docs search ImportCatalog

# Limit result count
b2c docs search authentication --limit 5

# List all available entries
b2c docs search --list

# List entries in a specific category
b2c docs search --list --category tooling

# List with workspace awareness — note that --list does NOT auto-detect; it narrows only when --workspace is specified
b2c docs search --list --workspace sfra

# Workspace awareness is ON by default — search auto-detects the project type
b2c docs search "components"

# Opt out of workspace awareness entirely
b2c docs search "components" --workspace all

# Force auto-detection explicitly
b2c docs search "hooks" --workspace auto

# Search with specific workspace types (boosts relevant docs)
b2c docs search "checkout flow" --workspace sfra

# Search for multiple workspace types (union of their boost categories)
b2c docs search "api integration" --workspace cartridges,pwa-kit-v3

# Bound the whole corpus to specific topics (an allowlist, not a per-query filter)
b2c docs search "login" --topics sfnext,commerce-api

# Or pin it via the environment (applies to every docs search/read in the shell)
export SFCC_DOCS_TOPICS=sfra,commerce-api,script-api,tooling
b2c docs search "checkout"

Search is content-aware — it indexes titles, section headings, summaries, and keywords. Developer Center guides have LLM-generated summaries and keywords for improved discoverability.

Restricting available topics (`--topics`)

--topics (or the SFCC_DOCS_TOPICS env var) is a launch-time allowlist that bounds the entire available corpus — distinct from the per-query --category filter and the soft --workspace boost. When set, only those categories are ever reachable; --category and --workspace narrow within the allowlist (their intersection wins), and docs read will not resolve an id outside it. Unknown category names are ignored with a warning. Use it to pin the docs surface to exactly the topics relevant to your project (for example, an SFRA team can drop PWA Kit and Storefront Next guides entirely). The equivalent for the MCP server is the --docs-topics startup flag.

Workspace-Aware Search

Workspace-aware search favors documentation relevant to your project type, and is on by default — docs search auto-detects the workspace framework unless you say otherwise:

Auto-detection (default, or --workspace auto): Detects the project type from the workspace structure and boosts relevant documentation categories
Framework-specific: Specify one or more workspace types comma-separated: cartridges, sfra, pwa-kit-v3, or storefront-next. Multiple types union their boost categories (never sum scores)
Disable awareness: Use --workspace all to search without any workspace preference

Category boost mapping:

Always relevant (any workspace): b2c-commerce, tooling
cartridges → script-api, job-step
sfra → sfra
pwa-kit-v3 → pwa-kit-managed-runtime, commerce-api
storefront-next → sfnext, commerce-api

SFRA vs cartridges distinction: A SFRA project is detected as both cartridges and sfra (SFRA is a refinement — it has the app_storefront_base cartridge). A cartridge project that is not SFRA (custom API cartridges, or a PWA Kit / Storefront Next repo that also ships cartridges) is cartridges only — so it will not boost sfra docs.

A detected or specified workspace always boosts relevant docs (ranks them higher) but never hides categories. To hard-scope results to a category, use --category or --topics.

Workspace awareness never filters or hides categories — it only reweights results by applying a x1.4 boost to relevant categories and a x0.3 de-boost to competing storefront frameworks (sfra, pwa-kit-managed-runtime, sfnext that are not the detected/specified workspace).

Output

Default output is a table with id, category, title, and score columns. Use --extended or --columns to show summary, keywords, url, and sourceUrl fields. With --list, output shows all entries with their metadata plus a total count.

b2c docs read

Read documentation for a specific entry. For online corpora (Script API, Developer Center guides, and Salesforce Help), content is fetched from the raw markdown source and cached locally after the first read (with a graceful offline fallback showing summary + headings + both URLs if the network fetch fails). Standard job steps are read from the bundled copy.

Usage

bash

b2c docs read <query>

Arguments

Argument	Description	Required
`query`	Entry ID, class/module name, or search query to match	Yes

Flags

Flag	Description	Default
`--raw`, `-r`	Output raw markdown (no terminal rendering)	`false`
`--topics`	Allowlist that bounds readable docs to these categories (comma-separated; env `SFCC_DOCS_TOPICS`). An id outside it will not resolve	(all)
`--workspace`	Favor a workspace type (`auto`\|`all`\|`cartridges`\|`sfra`\|`pwa-kit-v3`\|`storefront-next`) when resolving a fuzzy query, matching `docs search`. Ignored for an exact id match	`auto`

Examples

bash

# Read a Script API class doc
b2c docs read ProductMgr

# Read by fully qualified name
b2c docs read dw.catalog.ProductMgr

# Read a Developer Center guide by namespaced ID
b2c docs read sfnext/sfnext-get-started
b2c docs read commerce-api/slas-passwordless-login-registration

# Read a Salesforce Help article by namespaced ID
b2c docs read help-admin/b2c_site_import_export
b2c docs read help-merchant/b2c_creating_price_books

# Read a standard job step's purpose + configuration parameters
b2c docs read ImportCatalog

# Read the standard job step catalog overview
b2c docs read job-steps

# Read a tooling guide
b2c docs read guide-authentication

# Output raw markdown for piping
b2c docs read ProductMgr --raw

# JSON output with selected entry + content
b2c docs read ProductMgr --json

Output

By default, markdown is rendered for terminal display. Raw markdown is emitted when using --raw (or when output is not a TTY). For online corpora (Script API, Developer Center guides, Salesforce Help), the command fetches full content from the sourceUrl (.md) and caches it locally; if the fetch fails, it displays the locally-indexed summary, section headings, and both the url (.html) and sourceUrl (.md) so you can retrieve the page yourself.

Fuzzy read matches the search ranking: docs read resolves an exact id deterministically (e.g. dw.catalog.ProductMgr always reads that class). For a fuzzy query (e.g. productmgr), it applies the same workspace-aware ranking as docs search — auto-detecting the workspace by default, or honoring --workspace — so a fuzzy docs read resolves the same top hit that docs search ranks first. Use --workspace all to opt out of the preference.

b2c docs cache

Show or clear the local documentation content cache. Content for online corpora (Script API, Developer Center guides, Salesforce Help) is fetched on docs read and cached on disk under the CLI's cache directory (e.g. ~/.cache/b2c/docs-content on Linux, ~/Library/Caches/b2c/docs-content on macOS), so repeated reads — within a session and across CLI invocations — avoid the network for up to 7 days per entry.

Usage

bash

b2c docs cache [--clear]

Flags

Flag	Description	Default
`--clear`	Delete all cached documentation content (memory + on-disk)	`false`

Examples

bash

# Show cache location, file count, and total size
b2c docs cache

# Clear the cache (forces the next read to re-fetch from the source)
b2c docs cache --clear

# JSON output
b2c docs cache --json

Output

Without --clear, prints the cache directory, number of cached files, and total size. With --clear, removes both cache tiers and reports how many files (and how much space) were freed.

b2c docs schema

Read bundled XSD schemas (import/export data format definitions).

Usage

bash

b2c docs schema [query]

Arguments

Argument	Description	Required
`query`	Schema name or partial match (for example `catalog`, `order`)	No (required unless `--list` is used)

Flags

Flag	Description	Default
`--list`, `-l`	List all available schemas	`false`
`--path`, `-p`	Print the filesystem path to the schema instead of its content	`false`

Examples

bash

# Read a specific schema
b2c docs schema catalog

# Fuzzy match by schema name
b2c docs schema order

# List available schemas
b2c docs schema --list

# JSON output
b2c docs schema catalog --json

# Get the filesystem path to a schema
b2c docs schema catalog --path

Output

Without --json, the command writes schema XML directly to stdout. With --path, it prints the resolved filesystem path (useful for passing to XML validation tools). With --list, it prints available schema IDs and a total count.

Validating XML with xmllint

Use the --path flag to pass schema paths directly to xmllint for XML validation (requires installation of xmllint):

bash

xmllint --schema "$(b2c docs schema catalog --path)" catalog.xml --noout

b2c docs download

Download Script API documentation from a B2C Commerce instance to a local directory.

Usage

bash

b2c docs download <output>

Arguments

Argument	Description	Required
`output`	Local output directory for extracted docs	Yes

Flags

Flag	Description	Default
`--keep-archive`	Keep the downloaded archive file after extraction	`false`

In addition to global flags, this command supports instance flags and authentication flags for WebDAV access.

Examples

bash

# Download docs to a local directory
b2c docs download ./docs

# Keep the downloaded archive
b2c docs download ./docs --keep-archive

# Specify instance hostname directly
b2c docs download ./my-docs --server sandbox.demandware.net

# JSON output
b2c docs download ./docs --json

Output

The command reports the number of extracted files and output path. If --keep-archive is set, it also prints the saved archive location.

Docs Commands ​

Authentication ​

b2c docs search ​

Usage ​

Arguments ​

Flags ​

Examples ​

Restricting available topics (--topics) ​

Workspace-Aware Search ​

Output ​

b2c docs read ​

Usage ​

Arguments ​

Flags ​

Examples ​

Output ​

b2c docs cache ​

Usage ​

Flags ​

Examples ​

Output ​

b2c docs schema ​

Usage ​

Arguments ​

Flags ​

Examples ​

Output ​

Validating XML with xmllint ​

b2c docs download ​

Usage ​

Arguments ​

Flags ​

Examples ​

Output ​

Docs Commands

Authentication

b2c docs search

Usage

Arguments

Flags

Examples

Restricting available topics (`--topics`)

Workspace-Aware Search

Output

b2c docs read

Usage

Arguments

Flags

Examples

Output

b2c docs cache

Usage

Flags

Examples

Output

b2c docs schema

Usage

Arguments

Flags

Examples

Output

Validating XML with xmllint

b2c docs download

Usage

Arguments

Flags

Examples

Output