Slack API errors, explained one by one

The Web API signals failure in the body, not the status line: every response carries a boolean ok, and when it is false the error field holds a machine-readable string like channel_not_found. Only rate limiting changes the status code, to 429 with a Retry-After header. Any error handling built on HTTP status alone will treat every Slack failure as success.

channel_not_found means the channel value did not resolve — usually a #name passed where a C-prefixed ID is required, or a private channel the bot cannot see. not_in_channel means the channel resolved but the bot is not a member. The fix for the first is the right ID; the fix for the second is /invite or the chat:write.public scope for public channels.

The response tells you: needed names the scope the method requires and provided lists what your token has. Add the needed scope under OAuth & Permissions, then reinstall the app to the workspace — scopes only take effect on the token issued at install, so editing the scope list without reinstalling changes nothing.

Reinstalling or removing an app revokes previously issued tokens, and old tokens then return token_revoked (or invalid_auth). Tokens are also revoked when the installing user is deactivated on user-token installs. Always store the token issued by the most recent oauth.v2.access exchange and treat token_revoked as a signal to re-run the install flow.

ratelimited is the generic Web API answer, sent with HTTP 429 and a Retry-After header stating how many seconds to wait. rate_limited is documented on chat.postMessage for apps posting messages too quickly. app_rate_limited is an Events API event delivered when an app exceeds 30,000 event deliveries per workspace per hour — it arrives as an event, not as an API response.

Slack documents it as authentication that cannot be validated: the token is invalid — truncated, from another workspace, an app-level xapp- token used on a Web API method — or the request comes from an IP address the app or admin has disallowed. Check the token with auth.test first; if that passes, look at IP allowlists.

The app's Messages tab is switched off in its configuration, so Slack refuses bot DMs with messages_tab_disabled. Open the app's settings under App Home, enable the Messages Tab, and if you want users to write to the bot also tick 'Allow users to send Slash commands and messages from the messages tab'. No scope fixes this — it is an app configuration toggle.

On several validation failures — Block Kit problems in particular — Slack attaches a response_metadata object whose messages array contains human-readable strings pointing at the specific block or field that failed. It is the fastest way to locate an invalid_blocks failure in a 40-block payload, but only exists if you log the full response body rather than just the error field.

Yes. Node's @slack/web-api rejects the promise with an error whose code is ErrorCode.PlatformError and whose data property holds Slack's response body; rate-limited calls are retried automatically unless rejectRateLimitedCalls is true. Python's slack_sdk raises slack_sdk.errors.SlackApiError, with the string in e.response["error"] and Retry-After available via e.response.headers.

Slack requires an HTTP 2xx acknowledgment within 3 seconds of delivering an event. Miss it and Slack retries up to three times — almost immediately, after 1 minute, then after 5 minutes — marking each retry with X-Slack-Retry-Num and X-Slack-Retry-Reason headers. Acknowledge first and process asynchronously, and deduplicate on event_id.