Web Task
NPM · WEB-TASK-API · SCANNED SEP 20
Browser tasks, sessions, and recipes for MCP and HTTP automation.
Available components
How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →
Supply Chain Security88
- No malware found by supply-chain analysis.Pass
- CVE check failed: a known medium-severity CVE affects fastify 5.8.5, a direct dependency. A fixed version is available. View diagnostics → Fail
- No install/post-install scripts declared.Pass
- 44 of 145 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency6
- Repository check failed: the declared repository URL returned HTTP 404. See how to fix → View diagnostics → Fail
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- License check failed: no license is declared. See how to fix → Fail
- Actively maintained (last published 114 days ago).Pass
- Security-disclosure policy not yet verified: we couldn't inspect the source repository.Unverified
Schema Quality & AI Usability78
- AI-judged instruction clarity (excellent).Pass
- Tool/resource definitions use about 1143 tokens (~103/item across 11 items; 11 tools + 0 resources), lean.Pass
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management83
- Stability observed for 25 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage100
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 100% of tool parameters carry a description.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 11 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 11 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
How do I install the Web Task MCP server?
Web Task runs locally as an npm package, launched with npx -y web-task-api. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
npm · web-task-api
claude mcp add jojo-labs-web-task-api -- npx -y web-task-api
{
"mcpServers": {
"jojo-labs-web-task-api": {
"command": "npx",
"args": [
"-y",
"web-task-api"
]
}
}
} {
"servers": {
"jojo-labs-web-task-api": {
"command": "npx",
"args": [
"-y",
"web-task-api"
]
}
}
} codex mcp add jojo-labs-web-task-api -- npx -y web-task-api
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"jojo-labs-web-task-api": {
"type": "local",
"command": [
"npx",
"-y",
"web-task-api"
],
"enabled": true
}
}
} openclaw mcp add jojo-labs-web-task-api --command npx --arg -y --arg web-task-api
mcp_servers:
jojo-labs-web-task-api:
command: "npx"
args: ["-y", "web-task-api"] {
"McpServers": {
"jojo-labs-web-task-api": {
"Transport": "stdio",
"Command": "npx",
"Arguments": [
"-y",
"web-task-api"
]
}
}
} assistant mcp add jojo-labs-web-task-api -t stdio -c npx -a -y web-task-api
{
"mcpServers": {
"jojo-labs-web-task-api": {
"command": "npx",
"args": [
"-y",
"web-task-api"
]
}
}
} Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 19 Sept 26 −3
- Stability: pass → 0.80 functional
- 18 Sept 26 +1
- Stability: 0.97 → pass security
- 16 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.
- 14 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.
- 12 Sept 26 −3
- Stability: pass → 0.80 functional
- 11 Sept 26 +1
- Stability: 0.97 → pass security
- 9 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.
- 7 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 20 Sept 2026 · Analysed npm/web-task-api@0.10.1
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | npm |
Background: How many MCP packages publish verified provenance →
Vulnerabilities 2 findings
| ID | CVE | Severity | Vector | Fix available |
|---|---|---|---|---|
| GHSA-3m5p-2c4r-xxw2 | CVE-2026-16732 | medium | CVSS:3.1/AV:A/AC:H/PR:N/UI:N/S:C/C:H/I:N/A:N | yes |
| GHSA-w2qp-rph6-63g4 | CVE-2026-18504 | medium | CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:N | yes |
Background: What a vulnerability scan can and cannot prove →
Dependencies 145 packages
| Packages resolved | 145 |
|---|---|
| Stale | 43 |
| No linked repository | 1 |
| Tree resolution | Complete |
Background: SBOMs and build attestations, explained →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
webtask_auth_capture ~200
Open a real headed browser window, navigate to a URL, and wait for the user to log in and click the injected Capture button. Returns cleaned HTML with noise elements stripped. Use when the page requires login or returns empty/minimal content for automated fetches. Cookies and session state are stored in the named profile so subsequent smartfetch calls with the same profile reuse the authenticated session without re-logging in.
| Name | Type | Req | Description |
|---|---|---|---|
| profile | string | – | Named profile to persist the session under. Defaults to 'default'. The same name must be passed to smartfetch profile= to reuse the session. |
| timeout_ms | integer | – | How long to wait for the user to complete login and click Capture, in milliseconds. Defaults to 300000 (5 minutes). |
| url | string | yes | The URL to open in the browser window. |
| wait_for | string | – | Optional CSS selector to wait for before capturing (useful for SPAs that render content asynchronously). |
No output schema declared.
No examples provided.
webtask_create_session Create Web Session ~201
Create a reusable session for browser-state continuity across related tasks. Use this before multiple runs that should share login state, defaults, or recent context.
| Name | Type | Req | Description |
|---|---|---|---|
| agent_kind | string | – | Optional default planner backend for the session. |
| agent_model | string | – | Optional default planner model for the session. Auto, PI, and OpenCode use the catalog-backed selection when set. Defaults to openai/gpt-5.3-codex-spark. |
| agent_variant | string | – | Optional default planner thinking variant for the session. Defaults to high for catalog-backed planners. |
| mode | string | – | Session mode: guest for reusable browser state, profile for a named bound profile. |
| name | string | yes | Session name. |
| notes | string | – | Optional operator notes for the session. |
| profile | string | – | Optional bound browser profile for profile-mode sessions. |
| start_url | string | – | Optional default start URL for later tasks in this session. |
No output schema declared.
No examples provided.
webtask_get_session Get Web Session ~47
Fetch one saved session including recent task history. Use this to inspect continuity context before reusing a session in a new run.
| Name | Type | Req | Description |
|---|---|---|---|
| session_id | string | yes | Session id to fetch. |
No output schema declared.
No examples provided.
webtask_get_task Get Web Task ~60
Fetch one persisted task run with step history and artifact paths. Use this to inspect an earlier run instead of starting a new task; use list_tasks when you need to discover recent runs first.
| Name | Type | Req | Description |
|---|---|---|---|
| task_id | string | yes | Task id to fetch. |
No output schema declared.
No examples provided.
webtask_health Web Task Health ~36
Inspect runtime readiness, planner availability, and local inventory counts. Use this when you need to confirm the runtime is configured before starting work.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
webtask_list_models List Web Models ~25
List the catalog-backed planner models and supported thinking variants for web-task-api.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
webtask_list_recipes List Web Recipes ~35
List promoted starter recipes for common browser-task flows. Use this before a run when you want to see reusable guided paths.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
webtask_list_sessions List Web Sessions ~34
List saved sessions with browser-state continuity metadata. Use this when you need to find an existing session to reuse or inspect.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
webtask_list_tasks List Web Tasks ~75
List recent persisted task runs with optional status or session filters. Use this when you need to discover recent runs before fetching one task in detail.
| Name | Type | Req | Description |
|---|---|---|---|
| limit | integer | – | Maximum recent runs to return. |
| session_id | string | – | Optional session filter for related runs. |
| status | string | – | Optional task status filter. |
No output schema declared.
No examples provided.
webtask_run Run Web Task ~338
Start a new browser task from a goal and optional URL. Use this when you want the runtime to visit, read, or act on a site now; use session tools for continuity and get_task to inspect a saved run.
| Name | Type | Req | Description |
|---|---|---|---|
| agent_kind | string | – | Planner backend to use for this run. |
| agent_model | string | – | Optional planner model override for this run. Auto, PI, and OpenCode use the catalog-backed selection when set. Defaults to deepseek/deepseek-v4-flash. |
| agent_variant | string | – | Optional planner thinking variant override for this run. Defaults to high for catalog-backed planners. |
| goal | string | yes | What the browser task should accomplish right now. |
| headless | boolean | – | Whether to prefer headless browser mode for this run. |
| input | object | – | Structured task inputs that the recipe or planner can reference. |
| max_steps | integer | – | Maximum browser steps before the run stops. |
| mode | string | – | Task mode: read for extraction or research, act for interactive changes. |
| output_schema | object | – | Optional JSON Schema-like object used to validate the final structured result. |
| profile | string | – | Optional named browser profile for authenticated or warmed browser state. |
| recipe_id | string | – | Optional starter recipe id when you want a known flow instead of a purely freeform run. |
| session_id | string | – | Optional saved session to reuse when this task should continue earlier browser state or context. |
| start_url | string | – | Optional starting URL when you already know where the task should begin. |
| timeout_ms | integer | – | Overall run timeout in milliseconds. |
No output schema declared.
No examples provided.
webtask_update_session Update Web Session ~92
Update saved session metadata like name, profile, notes, or default start URL. Use this to adjust continuity defaults without creating a new session.
| Name | Type | Req | Description |
|---|---|---|---|
| name | string | – | Updated session name. |
| notes | string | – | Updated session notes. |
| profile | string | – | Updated bound profile. |
| session_id | string | yes | Session id to update. |
| start_url | string | – | Updated default start URL. |
No output schema declared.
No examples provided.
What is the Web Task MCP server?
Web Task is an MCP server listed in the public MCP registry as io.github.jojo-labs/web-task-api. Browser tasks, sessions, and recipes for MCP and HTTP automation. This page covers its npm package (web-task-api).
Is the Web Task MCP server safe to use?
Web Task scores 71 out of 100 on VerifyMCP. We recorded 2 known advisories against it as of 20 September 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.
What tools does the Web Task MCP server expose?
Web Task exposes 11 tools: webtask_run, webtask_get_task, webtask_list_tasks, webtask_list_recipes, webtask_list_models, and 6 more. Their descriptions and schemas cost roughly 1,143 tokens of context every time the server is loaded.
Is the Web Task MCP server still maintained?
Web Task is still listed as active in the MCP registry. We last reached this channel on 20 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.