Invalid API Request Breaks Claude Session During Anthropic Documentation Fetch

can you check what help settings claude code has available

⏺ Bash(claude code --help)…
  ⎿  Usage: claude [options] [command] [prompt]

Claude Code - starts an interactive session by default, use -p/--print
for
non-interactive output

Arguments:
prompt Your prompt

Options:
-d, --debug Enable debug mode
--verbose Override verbose mode setting from
config
-p, --print Print response and exit (useful for
pipes)
--json Output JSON in the shape {cost_usd:
number,
duration_ms: number, duration_api_ms:
number,
result: string} (only works with
--print)
--mcp-debug Enable MCP debug mode (shows MCP
server
errors)
--dangerously-skip-permissions Bypass all permission checks. Only
works in
Docker containers with no internet
access.
--allowedTools <tools...> Comma or space-separated list of tool
names
to allow (e.g. "Bash(git*) Edit
Replace")
-v, --version output the version number
-h, --help display help for command

Commands:
config Manage configuration (eg. claude
config set
-g theme dark)
mcp Configure and manage MCP servers
migrate-installer Migrate from global npm installation
to local
installation
doctor Check the health of your Claude Code
auto-updater
update Check for updates and install if
available

check what you can do in config

⏺ Bash(claude config --help)…
  ⎿  Usage: claude config [options] [command]

Manage configuration (eg. claude config set -g theme dark)

Options:
-h, --help display help for command

Commands:
get [options] <key> Get a config value
set [options] <key> <value> Set a config value
remove|rm [options] <key> [values...] Remove a config value or items
from a config array
list|ls [options] List all config values
add [options] <key> <values...> Add items to a config array
(space or comma separated)
help [command] display help for command

⏺ Bash(claude config list)…
  ⎿  {
"allowedTools": [],
"dontCrawlDirectory": false,
"hasTrustDialogAccepted": true,
"hasCompletedProjectOnboarding": true,
"ignorePatterns": []
}

i saw something about an api key

⏺ Let me check if there's information about API keys in the configuration
options:

⏺ Bash(claude config set --help)…
  ⎿  Usage: claude config set [options] <key> <value>

Set a config value

Options:
-g, --global Use global config
-h, --help display help for command

⏺ Fetch(url: "https://docs.anthropic.com/en/docs/agents-and-tools/claude-co…e
/overview", prompt: "Find any information about setting up API keys or
authentication for Claude Code")
⎿  Received 0 bytes (403 Forbidden)
Error: Request failed with status code 403
⎿  Received 0 bytes (403 Forbidden)
Error: Request failed with status code 403
⎿ API Error: 400 {"type":"error","error":{"type":"invalid_request_error","

hello

⎿ API Error: 400 {"type":"error","error":{"type":"invalid_request_error","

you there

⎿ API Error: 400 {"type":"error","error":{"type":"invalid_request_error","

/bug

  ⎿  Bug report submitted

Thanks for the report. Fix incoming!

Thanks for the report. Fix incoming!

having the same issue again please fix !

I pay for the 200 dollar MAX membership after a day of heavy use. I now get this message after the second or first reply I give claude.

⎿ API Error: 400
{"type":"error","error":{"type":"invalid_request_error","message":"tools.339.custom.name: String should
have at most 64 characters"}

RIGHT NOW CLAUDE CODE IS UNUSABLE

I'm hitting this one too. It happened in a big, complex conversation: I can't even compact it!

Here's my workaround:

1. Resume the conversation with claude --resume
2. Manually copy/paste the history into https://claude.ai/
3. Ask claude to summarize the convo as a markdown artifact
4. Copy/paste the markdown summary back into a new claude session

Wow this is really a bummer. Totally disrupts the... vibe, which is what generates the $ for Anthropic unless I'm mistaken. Please fix.

This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.