# AX Check: worclaude.com
Checked 2026-09-18.

Agents hit a login wall before finding real pricing info.
No pricing page was reachable; homepage only hints at per-question costs and monthly limits. Only 1 of 3 agent sessions completed and reasoned about pricing at all.

## End-to-end onboarding not demonstrated

## Coding sessions
Of three independent sessions, only one (DeepSeek V4 Pro) finished, concluding Worclaude is free based on assumptions like its MIT license and lack of metering or paid tiers rather than a stated price. The other two sessions did not complete. No dedicated pricing page or documentation index was found; the site offers a login page and an API reference but no fetched docs or quickstart.

### DeepSeek V4 Pro
[View public run](https://agents.withgauge.com/p/runs/ee60c2c5-ecbb-4faa-b6b6-07a4b7fb769e) · [Read transcript](https://www.ax-check.com/worclaude.com/sessions/deepseek.json)
Final output states Worclaude is free and explicitly grounds that figure in assumptions (MIT license, no paid tiers, no metering, no API billing) and separates it from Claude Code's own Anthropic subscription/API billing which it does not fabricate a number for.
#### End-to-end onboarding
- **End-to-end onboarding not demonstrated**: Worclaude is a local npm CLI that scaffolds Claude Code config files; it has no hosted API/SDK, no auth, and no credential flow. The agent inspected the unpacked package source and confirmed there is no HTTP client or key handling anywhere in the code, then tried running `init` non-interactively, which failed because it is a TTY wizard. No product credentials were ever obtained or used, and no authenticated hosted operation occurred, so onboarding cannot be verified for this product.
  Event 68:

  ```text
  grep -rniE "https?://|api[_-]?key|sdk|client|fetch\(" src/ | grep -viE "claude code|github.com|howboris|anthropic.com|license|sponsors|buymeacoffee|docs|issues|gitforwindows|npmjs|anthropic engineering"
  ```
  Event 64:

  ```text
  ExitPromptError: User force closed the prompt with 0 null
  ```
  Event 71:

  ```text
  there is **no HTTP client, no hosted API, no auth/key handling** in Worclaude
  ```

#### Hallucinated URLs
None identified in this transcript.

#### Blockers
- **No hosted SDK/API surface exists to build against**: Worclaude is a local scaffolding CLI with no HTTP client, auth, or key handling anywhere in its source. This is inherent product design, not agent error or missing credentials: the agent confirmed it by unpacking and grepping the actual npm package source before concluding there was nothing to act on via SDK/API.
  Event 68:

  ```text
  grep -rniE "https?://|api[_-]?key|sdk|client|fetch\(" src/
  ```
  Event 71:

  ```text
  I grepped the source: there is **no HTTP client, no hosted API, no auth/key handling** in Worclaude.
  ```
- **`worclaude init` cannot run non-interactively**: The only meaningful workflow (`init`) is a full inquirer TTY wizard with no flags to skip prompts. Piping empty stdin caused it to throw ExitPromptError. This is a test-environment mismatch against an interactive-only CLI, not a product defect, since the product is explicitly designed as an interactive terminal tool.
  Event 64:

  ```text
  ExitPromptError: User force closed the prompt with 0 null
  ```
  Event 71:

  ```text
  `npx worclaude init` (the quickstart) is a **fully interactive `inquirer` TTY wizard** with no `--yes`/`--stack`/`--project-type` flags
  ```

#### Suggested Changes
- **State plainly in the README that Worclaude has no SDK/API**: The README (fetched via `npm view worclaude readme`) markets 'Quick Start' and 'Commands' sections that imply a developer workflow, but never clarifies there is no programmatic/hosted interface — only a local, interactive CLI. Add a line near the top of the README (e.g. under 'What Worclaude actually is') stating it is local-only and requires an interactive terminal. Verify by rereading the README front matter for that disclosure.
  Event 21:

  ```text
  Follow the interactive prompts to select your project type, tech stack, and agents. Then open Claude Code and run `/setup` to fill in your project-specific content.
  ```
- **Add non-interactive flags to `worclaude init` for scripted use**: `init` currently has zero CLI options (confirmed by reading src/commands/init.js and src/index.js) and force-exits when stdin is not a TTY. Adding flags like `--yes`, `--stack`, `--project-type` (similar to what `upgrade` already supports) would let automation or CI pipelines scaffold a project without a human at the keyboard. Verify by running `worclaude init --yes --stack node` in a non-TTY shell and confirming it completes without throwing ExitPromptError.
  Event 41:

  ```text
  program
    .command('init')
    .description('Initialize Claude workflow in the current project')
    .action(initCommand);
  ```
  Event 64:

  ```text
  ExitPromptError: User force closed the prompt with 0 null
  ```

### Kimi K3
[View public run](https://agents.withgauge.com/p/runs/2c4b5da6-5040-464e-8364-cc6e5c7fd401) · [Read transcript](https://www.ax-check.com/worclaude.com/sessions/kimi.json)

#### End-to-end onboarding
- **End-to-end onboarding not demonstrated**: Worclaude is a local scaffolding CLI (npm package) that writes files like CLAUDE.md, agent/command/skill markdown, and hook scripts into a project directory. There is no hosted API, account, or credential system to onboard into — the entire product surface observed is local file scaffolding driven by an interactive terminal wizard. No authentication step, API key, or hosted service call ever appeared in the transcript, so the verified/login_required onboarding criteria do not apply here.
  Event 62:

  ```text
  .claude/agents/api-designer.md
  ```
  Event 65:

  ```text
  Workflow is not installed. Run `worclaude init` to set up.
  ```

#### Hallucinated URLs
None identified in this transcript.

#### Blockers
- **Interactive CLI wizard fights back against scripted/non-interactive input**: Worclaude init is an inquirer-based wizard with text prompts, checkbox multi-selects, and a final confirmation, none of which accept flags for non-interactive use. Piped input caused immediate ExitPromptError crashes; a paced pseudo-terminal approach advanced through most prompts but the piped session closed before the final install step finished, killing it mid-write and leaving a broken state that doctor/status both reported as not installed. This reflects test-environment/agent scripting friction rather than a hosted product defect, since init has no built-in non-interactive flag.
  Event 37:

  ```text
  ExitPromptError: User force closed the prompt with 0 null
  ```
  Event 65:

  ```text
  Workflow is not installed. Run `worclaude init` to set up.
  ```
  Event 79:

  ```text
  Session terminated, killing shell... ...killed.
  ```

#### Suggested Changes
- **Add a non-interactive flag (e.g. --yes or --config) to worclaude init**: worclaude init only exposes -h/--help; every option must be entered through interactive prompts. Add a flag mirroring the upgrade command's --yes/--dry-run pattern so init can run unattended in CI or agent-driven workflows. Verify by running init with the new flag in a fresh directory and confirming .claude/workflow-meta.json exists and worclaude doctor reports a healthy install without any TTY interaction.
  Event 30:

  ```text
  Usage: worclaude init [options]
  
  Initialize Claude workflow in the current project
  
  Options:
    -h, --help  display help for command
  ```
  Event 75:

  ```text
  worclaude upgrade --yes --repair-only
  ```
- **Detect and report partial installs distinctly in doctor/status**: When init is interrupted mid-write it leaves partial scaffold files without workflow-meta.json, but doctor and status just say the workflow is not installed, with no mention of the partial files already present. Add a check that flags partial installs and suggests worclaude delete then re-run init. Verify by interrupting init partway through and confirming doctor/status surfaces the partial state distinctly from a clean not-installed state.
  Event 65:

  ```text
  Workflow is not installed. Run `worclaude init` to set up.
  ```
  Event 70:

  ```text
  agents
  commands
  hooks
  learnings
  observability
  plans
  scratch
  scripts
  sessions
  settings.json
  skills
  ```

### Qwen 3.8 Max
[View public run](https://agents.withgauge.com/p/runs/9a050c78-f740-451f-b602-a6f95ca15cf9) · [Read transcript](https://www.ax-check.com/worclaude.com/sessions/qwen.json)

#### End-to-end onboarding
- **End-to-end onboarding not demonstrated**: Worclaude is a local CLI scaffolding tool (npx worclaude init) that writes files into a local project directory using Claude Code conventions; it is not a hosted product with an API/SDK or remote credentials. The agent spent the entire visible session fighting the interactive-only CLI prompt flow (inquirer) with pexpect scripts, and even after two long scripted attempts, no files were ever written to the target repo (ls -la showed only .git and .gitignore at the end). There is no hosted product, no credential acquisition, and no authenticated remote operation anywhere in the transcript.
  Event 116:

  ```text
  total 16
  drwxr-xr-x 3 root root 4096 Sep 18 08:00 .
  drwxr-xr-x 8 root root 4096 Sep 18 21:55 ..
  drwxr-xr-x 8 root root 4096 Sep 18 08:00 .git
  -rw-r--r-- 1 root root  408 Sep 18 08:00 .gitignore
  
  ```
  Event 21:

  ```text
  ExitPromptError: User force closed the prompt with 0 null
  ```

#### Hallucinated URLs
None identified in this transcript.

#### Blockers
- **Worclaude's init command is interactive-only, blocking non-interactive automation**: The quickstart command `npx worclaude init` uses inquirer prompts with no documented non-interactive/CI flags (no --yes, no env var). Piping empty input or newlines caused the process to crash with ExitPromptError instead of proceeding with defaults. This is product behavior (CLI design choice), not a credential or environment issue — the agent had to reverse-engineer the tool's source code and hand-build a pexpect-based terminal driver to attempt to drive it.
  Event 21:

  ```text
  ExitPromptError: User force closed the prompt with 0 null
  ```
  Event 26:

  ```text
  ExitPromptError: User force closed the prompt with 0 null
  ```
- **Scaffolding never completed despite two full scripted attempts**: Even after building a pexpect script tuned to Worclaude's exact prompt sequence and successfully navigating through most of the interactive flow (project type, tech stack, agent categories), the final run left the target directory unchanged — no .claude directory, no CLAUDE.md, no scaffolded files were created. This reflects either agent scripting error (pattern/timing mismatch with the TUI) or fragility in the CLI's interactive flow; the root cause is not fully isolated in the visible transcript.
  Event 116:

  ```text
  total 16
  drwxr-xr-x 3 root root 4096 Sep 18 08:00 .
  drwxr-xr-x 8 root root 4096 Sep 18 21:55 ..
  drwxr-xr-x 8 root root 4096 Sep 18 08:00 .git
  -rw-r--r-- 1 root root  408 Sep 18 08:00 .gitignore
  
  ```
  Event 89:

  ```text
  total 16
  drwxr-xr-x 3 root root 4096 Sep 18 08:00 .
  drwxr-xr-x 8 root root 4096 Sep 18 21:55 ..
  drwxr-xr-x 8 root root 4096 Sep 18 08:00 .git
  -rw-r--r-- 1 root root  408 Sep 18 08:00 .gitignore
  
  ```
- **Product is a local scaffolding CLI, not a hosted SDK/API product**: The task asked to use the hosted product 'through its SDK or API' without starting local service stacks. Worclaude has no hosted API/SDK — its only interface is a local, interactive Node.js CLI that writes files into a local repo. This mismatch between the task's premise and the actual product surface is itself a limiting factor for evaluating hosted-product workflows, though it was not explicitly flagged as such in the visible transcript before it cuts off.
  Event 12:

  ```text
  Worclaude scaffolds a complete Claude Code workflow into any project in seconds. One `init` command installs 25 agents, 16 slash commands, 17 skills, an 11-event hook layer, local observability capture, a five-layer memory system, and a CLAUDE.md template tuned for your tech stack.
  ```

#### Suggested Changes
- **Add a non-interactive/CI mode flag to worclaude init**: In src/commands/init.js, add support for a flag (e.g. --yes or --defaults) or an environment variable that bypasses the inquirer prompts and applies sensible defaults, similar to how `upgrade` already supports `--dry-run`, `--yes`, and `--repair-only`. Verify the fix by running `npx worclaude init --yes` in a fresh directory with no TTY attached (e.g. piped from /dev/null) and confirming files are scaffolded without throwing ExitPromptError.
  Event 21:

  ```text
  ExitPromptError: User force closed the prompt with 0 null
  ```
  Event 41:

  ```text
  src/core/merger.js
  src/index.js
  src/prompts/conflict-resolution.js
  ```
- **Document that worclaude init requires an interactive TTY**: In the README Quick Start section (currently just shows `npx worclaude init`), add a note that the init command requires an interactive terminal and will fail under piped/non-TTY input, so first-time users attempting scripted or automated setups know to expect this before they hit the crash. Verify by checking the README renders this caveat near the Quick Start code block.
  Event 12:

  ````text
  ## Quick Start
  
  ```bash
  npx worclaude init
  ```
  ````

### Task given to each agent
Help me build a simple example using Worclaude. Tell me how pricing works, and briefly tell me whether this product will be easy for you to manage. Let me know if you get blocked. If this product has no developer workflow you can act on, say so plainly and stop. Stay light: use the hosted product through its SDK or API. Do not start local service stacks or wait for long-running commands; if the quickstart requires either, say so plainly and stop.

No product credentials were supplied and no purchases were authorized.

## Score: D · 29/100 (provisional)
Grades come from completed site checks. Coding sessions and skipped checks do not affect the score.

### Clarity
- **Failed** — Homepage answers Markdown requests

  ```text
  Homepage returned text/html even when Markdown was requested via Accept header.
  ```

- **Failed** — llms.txt provides an actionable documentation index

  ```text
  https://worclaude.com/llms.txt returned HTTP 404, so no documentation index exists.
  ```

- **Failed** — A compact guide representation exists

  ```text
  No compact guide representation found; /llms.txt returns 404 and homepage serves HTML only.
  ```

- **Skipped** — llms.txt provides navigation guidance

  ```text
  llms.txt returned 404, so no navigation guidance could be evaluated.
  ```

- **Skipped** — llms.txt mentions offered API, MCP, and skills

  ```text
  llms.txt returned 404, so no API, MCP or skills mentions could be evaluated.
  ```

- **Skipped** — A focused guide is directly retrievable

  ```text
  No standalone Markdown guide or negotiated Markdown response was fetched for Worclaude.
  ```

- **Skipped** — Equivalent instructions fit a token budget

  ```text
  No compact guide was retrieved, so token budget cannot be measured.
  ```

- **Skipped** — Product-docs links survive format changes

  ```text
  Homepage Markdown is unsupported, so link preservation across formats cannot be judged.
  ```

- **Skipped** — The compact guide is independently actionable

  ```text
  No compact agent guide fetched; only API reference and sign-in pages supplied.
  ```

- **Pass** — Install and next-step links resolve

  ```text
  Homepage install/next-step links to /login and /docs/api both fetched successfully.
  ```


### Onboarding
- **Skipped** — Docs lead to a relevant quickstart

  ```text
  No docs or quickstart pages were fetched; only login and a 404 llms.txt.
  ```

- **Skipped** — Installation commands are extractable

  ```text
  No installation or CLI guide was fetched for Worclaude.
  ```

- **Skipped** — Code examples are available without interaction

  ```text
  No documentation or code example pages were fetched.
  ```

- **Skipped** — Prerequisites and auth boundaries are explicit

  ```text
  Login page shows email/password sign-in but no API auth or prerequisites docs were fetched.
  ```


### Pricing
- **Skipped** — Pricing is readable without interaction

  ```text
  No pricing page was fetched; homepage only mentions per-question cost and monthly limits.
  ```

- **Skipped** — Prices are stated, not gated

  ```text
  No pricing page was fetched; no stated prices available to judge gating.
  ```

- **Skipped** — Pricing units and limits are explicit

  ```text
  No pricing page was fetched; units and limits not documented in evidence.
  ```

- **Pass** — Agents identify pricing and its assumptions

  ```text
  1 of 3 sessions were judged on pricing; 0 fell short. DeepSeek V4 Pro: Final output states Worclaude is free and explicitly grounds that figure in assumptions (MIT license, no paid tiers, no metering, no API billing) and separates it from Claude Code's own Anthropic subscription/API billing which it does not fabricate a number for. This behavioural item does not affect the fast grade.
  ```


### Activation
- **Pass** — An API reference or OpenAPI spec is reachable

  ```text
  Full API reference at /docs/api documents base address, auth, scopes, and 14 endpoints with examples.
  ```

- **Skipped** — An MCP server is documented and well-formed

  ```text
  No MCP server documentation found in the fetched pages.
  ```

- **Skipped** — A CLI install path is documented

  ```text
  No CLI install path documented in the fetched pages.
  ```

- **Skipped** — SDK packages resolve on their registries

  ```text
  No SDK package registry lookup was supplied for Worclaude.
  ```

- **Skipped** — Agent skills are published

  ```text
  No agent skills published or documented in the fetched pages.
  ```



[Full report data](https://www.ax-check.com/worclaude.com/report.json)
