Atlas / Skills / pytorch / Triaging Issues

Triaging IssuesCAUTION

skills/pytorch/triaging-issues

Agent skill by pytorch · autograd · deep learning · gpu · Audited 2026-10-28

⚑ Report / re-scan Source ↗

Tensors and Dynamic neural networks in Python with strong GPU acceleration

Verdict

CAUTION

Grade

Trust score

89 /100

Version

—

Hosts

—

License

NOASSERTION

Stars

105,132

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

Summary

This is a skill for auto-triaging issues. This is the human side of things :) There are 4 pieces to this skill;

SKILL.md this is the main description of what to do/ directions to follow. If you notice a weird anti pattern in triaging

this is the file you should update. The basic workflow is that there is a static list of labels in labels.json that the agent will read w/ their descriptions in order to make decisions. NOTE This is static and if new labels are added to pytorch/pytorch we should bump this list w/ their description. I made this static because the full set of labels is too big/quite stale. And I wanted to add more color to certain descriptions. The mechanics for actually interacting w/ gh issues is through the official mcp server. For V1, we always apply bot-triaged whenever any triage action is taken; you can filter those decisions here: https://fburl.com/pt-bot-triaged

templates.json: This is basically where we want to put canned responses. It includes redirect_to_forum (for usage questions) and

request_more_info (when classification is unclear). There are likely others we should add here as we notice more patterns.

There are hooks in /scripts: a target pre-hook (validate_issue_target.py) that blocks mutations outside the repository and issue selected by the workflow, a label pre-hook (validate_labels.py) that filters out labels we never want the bot to add, and a post-hook (add_bot_triaged.py) that automatically applies bot-triaged after any issue mutation.
The gh action uses a two-stage workflow to support issues opened by OSS users:
Stage 1 (.github/workflows/claude-issue-triage.yml): Triggers on issues: opened, captures the issue number, and uploads it as an artifact. This stage has no protected environment, so OSS actors can run it.
Stage 2 (.github/workflows/claude-issue-triage-run.yml): Triggers on workflow_run completion of Stage 1. Runs in the protected bedrock e

Read from source at commit aeb350f40dccOBSERVED · 2026-10-28

What it tells the agent

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: triaging-issues
description: Triages GitHub issues by routing to oncall teams, applying labels, and closing questions. Use when processing new PyTorch issues or when asked to triage an issue.
hooks:
  PreToolUse:
    - matcher: "mcp__github__issue_write|mcp__github__update_issue|mcp__github__add_issue_comment|mcp__github__transfer_issue"
      hooks:
        - type: command
          command: "python3 \"$CLAUDE_PROJECT_DIR\"/.agents/skills/triaging-issues/scripts/validate_issue_target.py"
    - matcher: "mcp__github__issue_write|mcp__github__update_issue"
      hooks:
        - type: command
          command: "python3 \"$CLAUDE_PROJECT_DIR\"/.agents/skills/triaging-issues/scripts/validate_labels.py"
  PostToolUse:
    - matcher: "mcp__github__issue_write|mcp__github__update_issue|mcp__github__add_issue_comment|mcp__github__transfer_issue"
      hooks:
        - type: command
          command: "python3 \"$CLAUDE_PROJECT_DIR\"/.agents/skills/triaging-issues/scripts/add_bot_triaged.py"
---

# PyTorch Issue Triage Skill

This skill helps triage GitHub issues by routing issues, applying labels, and leaving first-line responses.

## Contents
- [MCP Tools Available](#mcp-tools-available)
- [Labels You Must NEVER Add](#labels-you-must-never-add)
- [Issue Triage Steps](#issue-triage-for-each-issue)
  - Step 0: Already Routed — SKIP
  - Step 1: Question vs Bug/Feature
  - Step 1.5: Needs Reproduction — External Files
  - Step 2: Transfer
  - Step 2.5: PT2 Issues — Special Handling
  - Step 3: Redirect to Secondary Oncall
  - Step 4: Label the Issue
  - Step 5: Escalate — High Priority (human review), then release triage
  - Step 6: bot-triaged (automatic)
  - Step 7: Mark Triaged
- [V1 Constraints](#v1-constraints)

**Labels reference:** See [labels.json](labels.json) for the full catalog of labels suitable for triage. **ONLY apply labels that exist in this file.** Do not invent or guess label names. This file excludes CI triggers, test configs, release notes, deprecated labels, and labels requiring human decision.

**PT2 triage guide:** See [pt2-triage-rubric.md](pt2-triage-rubric.md) for detailed labeling guidance when triaging PT2/torch.compile issues.

**Response templates:** See [templates.json](templates.json) for standard response messages.

---

## MCP Tools Available

Use these GitHub MCP tools for triage:

| Tool | Purpose |
|------|---------|
| `mcp__github__get_issue` | Get issue details and existing labels |
| `mcp__github__get_issue_comments` | Get existing issue comments |
| `mcp__github__update_issue` | Apply labels or close issues |
| `mcp__github__add_issue_comment` | Add comment (only for redirecting questions) |
| `mcp__github__search_issues` | Find similar issues for context |

---

## Labels You Must NEVER Add

| Prefix/Category | Reason |
|-----------------|--------|
| Labels not in `labels.json` | Only apply labels that exist in the allowlist |
| `ciflow/*` | CI job triggers for PRs only |
| `test-config/*` | Test suite selectors for PRs only |
| `release notes: *` | Auto-assigned for release notes |
| `ci-*`, `ci:*` | CI infrastructure controls |
| `sev*` | Severity labels require human decision |
| `merge blocking` | Requires human decision |
| `actionable`, `needs design`, `needs reproduction`, `needs research` | Reserved for human reviewers after they have reviewed the issue |
| Any label containing "deprecated" | Obsolete |
| `oncall: releng` | Not a triage redirect target. Use `module: ci` instead |

**If blocked:** When a label is blocked by the hook, add ONLY `triage review` and stop. A human will handle it.

These rules are enforced by a PreToolUse hook that validates all labels against `labels.json`.

### Never Override Human Labels

If a human has already applied labels (especially `ci: sev`, severity labels, or priority labels), do NOT remove or replace them. Your job is to supplement, not override.

---

## Issue Triage (for each issue)

### 0) Already Routed — SKIP

**If an issue already has ANY `oncall:` label, SKIP IT entirely.** Do not:
- Add any labels
- Add `triaged`
- Leave comments
- Do any triage work

That issue belongs to the sub-oncall team. They own their queue.

### 1) Question vs Bug/Feature

- If it is a question (not a bug report or feature request): close and use the `redirect_to_forum` template from `templates.json`.
- If unclear whether it is a bug/feature vs a question: request additional information using the `request_more_info` template and stop.

### 1.5) External Files

Check if the issue body contains links to external files that users would need to download to reproduce.

**Patterns to detect:**
- File attachments: `.zip`, `.pt`, `.pth`, `.pkl`, `.safetensors`, `.onnx`, `.bin` files
- External storage: Google Drive, Dropbox, OneDrive, Mega, WeTransfer links
- Model hubs: Hugging Face Hub links to model files

**Action:**
1. **Edit the issue body** to remove/redact the download links
   - Replace with: `[Link removed - external file downloads are not permitted for security reasons]`
2. Use the `request_self_contained_reproduction` template from `templates.json`
3. Do NOT add `triaged` — wait for the user to provide a reproducible example

### 1.55) Missing Reproduction — Other Cases

Request a self-contained reproduction and stop when:
- The user reports a hardware-specific issue (e.g., specific GPU model) without a self-contained repro script
- The user references a specific model/checkpoint/dataset that is not publicly runnable in a few lines
- The issue describes version-upgrade breakage but only provides a high-level description without a minimal script
- The repro depends on a specific training setup, distributed environment, or non-trivial infrastructure

### 1.6) Edge Cases & Numerical Accuracy

If the issue involves extremal values or numerical precision differences:

**Patterns to detect:**
- Values near `torch.finfo(dtype).max` or `torch.finfo(dtype).min`
- NaN/Inf appearing in outputs from valid (but extreme) inp

Trust audit

CAUTIONgrade B · trust 89/100 Install with care. The audit found things worth knowing before you trust its output.

Layer	What it checks	Result
L0	Provenance & inventory	WARN
L1	Static analysis of the code	PASS
L2	Instruction surface (what it tells the agent)	PASS
L3	Class-specific surface	PASS
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: none-observed
Network: none-observed
Shell: none-observed
Dependencies: pinned
Secrets in source: none-found