PyTorch Issue Triage
Structures PyTorch GitHub issues, routes them, applies allowed labels, and records triage decisions consistently.
- Skill Road
- PyTorch Issue Triage
Categories
PyTorch Issue Triage is an official skill in the PyTorch repository. It helps maintainers and on-call teams review GitHub issues through a defined sequence, identify the responsible area, choose permitted labels, and prepare an appropriate next step. According to the provider, the skill is intended for processing new PyTorch issues and covers routing, requests for more information, label selection, and recording completed triage. It is not a ticketing system, an autonomous root-cause analyzer, or a promise that a single model response can replace a maintainer’s judgment.
Workflow and input context
The process starts with the complete issue and its existing labels. According to the provider, an issue carrying any on-call label is skipped because it already belongs to a responsible queue. For questions that are not bug reports or feature requests, the workflow calls for a forum redirect and closure with an approved response template. When the situation is ambiguous, the next step is to request additional information. The skill also distinguishes external files and cases that cannot yet be reproduced. Links to downloadable files or model artifacts should be removed for security reasons, while the reporter is asked for a self-contained reproduction. A hardware-specific problem, a model that cannot be run publicly, or a complex distributed setup may likewise require a reproduction before further classification.
Routing and domain labels
For issues belonging to another PyTorch project, the skill describes transferring the issue to the appropriate repository. It lists specialized on-call destinations for areas such as JIT, Distributed, Export, Quantization, Mobile, Profiler, and Visualization. PT2 is explicitly handled differently: the oncall: pt2 label does not end general triage; it requires a suitable module label and further steps. The documentation calls out common routing mistakes. MPS is not Mobile, DTensor belongs with Distributed, and ONNX receives a module label rather than an unsupported on-call destination. The affected code area should be inferred by asking where a fix would need to be made, not by reacting to one keyword in a stack trace or issue title.
Safety and decision boundaries
According to the provider, only labels present in the label reference may be applied. CI controls, deprecated labels, human-reserved severity labels, and unsupported on-call names must not be invented or added. Human-applied labels are not overridden. For segfaults, silent correctness problems, regressions, internal assertions, or issues affecting many users, the workflow first calls for triage review and human confirmation; high priority must not be assigned independently. The skill also separates security, privacy, and data-exposure risks from ordinary routing work. Updating issues, adding comments, transferring issues, and closing them are external actions that require checking the repository, issue number, permissions, and current context before execution. The supplied hooks validate targets and labels, but they do not replace accountable human review.
Practical value and limitations
The skill is useful as a consistent checklist for a large open-source project with many modules and specialized teams. It can help new maintainers avoid missing established routing rules and preserve the intended order of decisions. It only knows the information available through the connected GitHub context, however. Similar issues, reproductions, external files, and current ownership must actually be checked. Labels can change, so labels.json and the other reference files in the same source revision remain authoritative. PyTorch publishes its source under the BSD-3-Clause license. This entry was reviewed on September 9, 2026 against the official skill file and the official PyTorch documentation site. It describes the provider’s stated scope, not a guarantee of correct triage, availability, or security in every environment. GitHub stars are not stored because the Skill model has no github_stars field.
Installation instructions
What the PyTorch Issue-Triage skill is
The triaging-issues skill lives in the official PyTorch repository on GitHub, in the directory for Claude skills, and according to its description it is meant to triage newly incoming GitHub issues in the PyTorch project by routing them to the responsible on-call teams, applying appropriate labels, and, for pure questions, leaving a first-line reply and closing the issue. PyTorch is one of the most widely used open-source machine learning libraries and correspondingly receives a large number of issues daily, so a structured, partially automated first pass at sorting is, according to the project, useful for relieving human on-call teams and answering questions faster.
The triage flow in several steps
The flow described in the skill documentation is organized into several sequential steps. First it checks whether an issue has already been processed before, in which case the process is skipped. This is followed by distinguishing between a pure question and an actual bug or feature report, then checking whether external files are needed for a possible reproduction. The issue can then be transferred to a different team if needed, with a separate, more detailed rulebook file specifically for topics around PT2, i.e. torch.compile, containing specific labeling guidance. This is followed by a possible redirect to a secondary on-call team, the actual labeling of the issue with matching labels, an escalation tier for especially urgent cases requiring human review, and finally an automatic mark as bot-triaged and a final mark as triaged.
Technical safeguards via hooks and a fixed label list
A central safety feature of the skill is a strictly bounded list of allowed labels, stored in a separate file, which according to the documentation deliberately excludes CI trigger labels, test-config labels, release-note labels, deprecated labels, and labels that require a human decision. Per explicit instruction, only labels from this list may be used; inventing or guessing new label names is prohibited. The skill is additionally secured technically via so-called hooks that run automatically before and after every write-capable GitHub call, such as editing an issue, adding a comment, or transferring an issue to another repository. Before the actual action, a script validates the target of the operation, a second script checks the labels to be set against the allowed list, and after the action a third script automatically records that the issue was triaged by the bot.
Step 1: Available tools and prerequisites
For the actual interaction with GitHub, the skill has several MCP tools available according to the documentation, which can retrieve issue details and existing labels, view existing comments, set labels or close issues, add new comments, and search for similar issues to provide better context. In addition, pre-written response templates are available in a separate file, used for recurring situations such as redirecting questions. Using the skill therefore requires an agent environment with access to a GitHub MCP server and write permissions in the PyTorch repository, which are typically granted only to a restricted, trusted bot account.
Limitations, security, and best practices
The documentation explicitly notes limitations of the first version, which deliberately keep the feature scope narrow to avoid misbehavior such as applying incorrect labels or unnecessary issue transfers. The combination of a fixed label allowlist, mandatory validation hooks before every write action, and automatic tracking of which issues the bot has already processed serves to significantly reduce the risk of faulty or duplicate processing. Nevertheless, the skill remains a partially automated tool that explicitly escalates especially urgent cases for human review rather than deciding them autonomously. Other projects can adopt the general structure of a rulebook, a vetted label list, and validation hooks as a template, but the concrete content such as labels and team assignments is tightly tailored to the PyTorch project and cannot be transferred to other repositories without adaptation.
- Provider
- PyTorch
- License
- BSD-3-Clause
- Last reviewed
- 09.09.2026
Repository and documentation
Categories
Compatible with
Related guides
Guides and background related to this entry.