菜单
Debug a failed or misbehaving dltHub Platform deployment. Use when a runtime job fails, produces unexpected results, or the user wants to check job status and logs.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Build and deploy a minimal custom REST API pipeline to dltHub Platform. Use when the user says "Help me build and deploy a minimal pipeline", "I've finished onboarding, now what?", or "I just ran dlthub-start and want to build my own pipeline".
Deploy and run the pre-shipped Jaffle Shop sample pipeline on dltHub Platform — the final onboarding step after uvx dlthub-start. Use when the user says 'deploy the pipeline', 'deploy the pre-built pipeline', 'deploy the onboarding pipeline', 'do the onboarding task', 'finish the onboarding task', or is ready to complete onboarding. Assumes scaffolding, login, and playground workspace connection are already done.
The guided entry point for dltHub workbench use cases — ingestion from APIs, data exploration, transformations, deployment, data quality. Use when the user names a use case or wants to be oriented before starting: 'I want to ingest from Stripe', 'show me how to go from data to dashboard', 'take me through the full workflow', 'explore the workbench', 'what can I do with dlthub', 'give me a quick start', 'show me a demo', 'walk me through ingestion to visualization', 'I want to try everything end-to-end', 'teach me dltHub'. Do NOT use when the user is asking what's available or where to start in general — use the `dlthub-router` skill (in init) for capability-discovery questions ('what can you do', 'what toolkits are there', 'I'm new to dlthub'). Do NOT use when the user already has a specific task underway (debugging, adding an endpoint, deploying).
Create trigger evaluation setup for a toolkit skill. Use when the user wants to test whether a skill's description triggers correctly, set up eval workspaces, or generate trigger test queries for a skill. Use when user says 'create eval', 'test triggers', 'eval skill', or wants to measure skill triggering accuracy.
Run trigger evaluation for a skill and analyze results. Use when the user wants to run evals, check trigger accuracy, analyze clashes, or improve a skill description based on eval results. Use when user says 'run eval', 'test triggers', 'check skill triggering', or 'analyze eval results'.
The entry point for building anything with dlthub. Use this skill to route the user to the right workflow toolkit and install it on demand. MUST use when the user asks 'what can you do', 'what can I build', 'what are toolkits', 'how do I build a pipeline', 'I want to pull data from a REST API', 'ingest from a SQL database', 'load CSVs from S3', 'make reports / dashboards', 'transform / model my data', 'add data quality checks', 'how do I deploy / schedule a pipeline', 'I'm new to dlthub', 'where do I start', or seems unsure what to do next after setup. Also use whenever the user expresses a data-engineering goal but no matching workflow toolkit is installed yet — this skill installs it on demand. Do NOT use when the toolkit matching the user's intent is already installed — go straight to its entry skill instead; only route/install when the matching toolkit is missing. Do NOT use when a specific task is already in progress (debugging a pipeline, validating data, adding endpoints) and its toolkit is installed.
| name | debug-deployment |
| description | Debug a failed or misbehaving dltHub Platform deployment. Use when a runtime job fails, produces unexpected results, or the user wants to check job status and logs. |
Reference: https://dlthub.com/docs/hub/pipeline-operations/monitoring.md (pipeline health, logs, failure diagnosis)
Commands accept job names or selectors (fnmatch patterns):
dlthub job list # all jobs
dlthub job list batch # only batch jobs
dlthub job list "tag:ingest" # jobs tagged "ingest"
dlthub job list "schedule:*" # jobs with a schedule trigger
dlthub job info <name> # details for one job
dlthub job runs list [name_or_selector] # runs for matching jobs
dlthub job runs list [name_or_selector] --running # only active runs
dlthub job runs info <name> [run#] # specific run details
dlthub deploy --dry-run # preview manifest reconciliation
dlthub deploy --show-manifest # dump full manifest as YAML
dlthub job logs <name_or_selector> # latest run
dlthub job runs logs <name> [run#] # specific run
dlthub job logs <name> -f # stream in real-time
dlthub job cancel <name> # cancel active runs for one job
dlthub job cancel "tag:backfill" # cancel by selector
dlthub job cancel batch --dry-run # preview what would be cancelled
dlthub job runs cancel <name> [run#] # cancel a specific run
configured (list profiles). if not: prod profile (if configured)to run a single command on given profile use:
dlthub local run my_pipeline.py --profile prod # run under prod profile
WORKSPACE__PROFILE=prod uv run dlthub local pipeline info my_pipeline # env var for dlthub commands
Note: you must pin the production profile for mcp server to see the change
dlthub job trigger <selector> # trigger jobs by selector (e.g. tag:backfill)
dlthub job trigger <selector> --refresh # trigger with refresh signal
dlthub job trigger <selector> --profile <name> # trigger under a specific profile (e.g. prod)
dlthub job trigger <selector> --dry-run # preview which jobs would fire
dlthub pipeline run <pipeline_name> # trigger job by pipeline name
dlthub workspace connect <name_or_id> # switch workspace without re-login
dlthub info # workspace deployment overview
dlthub workspace deployment list # deployment version history
dlthub workspace deployment info [version] # details for a deployment version
dlthub workspace deployment sync [version] [--dry-run] # sync local files to remote (creates new deployment)
dlthub workspace configuration list # configuration version history
dlthub workspace configuration info [version] # details for a configuration version
dlthub workspace configuration sync [version] [--dry-run] # sync local config files to remote
dlthub local clean # clean local workspace state
dlthub local clean --skip-data-dir # clean but keep data directory
dlthub local pipeline list # all local pipelines
dlthub local pipeline info [pipeline_name] # local pipeline details
dlthub local pipeline failed-jobs [pipeline_name] # list failed load packages
dlthub local pipeline trace [pipeline_name] # show last trace
dlthub show
Prints the dltHub web UI URL. It should open automatically, but if the user says it does not, ask them to open it themselves.
If a job failed:
dlthub job runs info <name> [run#] -- check exit status and timingdlthub job runs logs <name> [run#] -- read the error outputpyproject.toml -- all packages must be declared, not just locally installedprod profile -- runtime uses prod profile, ask the user to check .dlt/prod.secrets.toml — NEVER access it directly, only the user may modify itif __name__ == "__main__": -- the job does nothing without itdev_mode=True left in -- drops and recreates dataset on every runexecute={"timeout": "6h"} in the decoratordlthub run <name_or_file>