Read a file from a deployed site

Return the bytes of one file currently served by the site. Use this to inspect or edit existing content (call read_file → modify → update_site mode:'patch') so

liveunknown~5s typical

mcpvibedeploy

$0.0020

per call

plus $0.001 platform fee per run

Open MCP server; Cracked bills routing only

Provider Upstream docs

$Set up https://cracked-ai-pearl.vercel.app/SKILL.md, then use Cracked to run mcp-be-vibedeploy-vibedeploy /read_file for me.

Input

Field	Type	Description
name	string	Site name (subdomain) or custom domain. Same lookup rules as get_site.
siteId	string	Alternative to name. One of name\|siteId is required.
path*	string	Site-relative path of the file to read (e.g. 'index.html', 'assets/main.css'). No leading
maxBytes	integer	Per-file size cap. Default 1048576, hard max 5242880. If the file is larger, the call fail

Call it

curl

curl https://cracked-ai-pearl.vercel.app/v1/run \
  -H "Authorization: Bearer ck_live_..." -H "content-type: application/json" \
  -d '{"provider":"mcp-be-vibedeploy-vibedeploy","endpoint":"/read_file","input":{"path":"..."}}'

cli

npx cracked-ai run -p mcp-be-vibedeploy-vibedeploy -e /read_file -i '{"path":"..."}'

mcp

run_tool({ provider: "mcp-be-vibedeploy-vibedeploy", endpoint: "/read_file", input: {"path":"..."} })

Try it

Runs against your signed-in workspace balance. Sign in if you have not.