name: Bug report
description: A skill or repository tool behaves incorrectly — wrong output, broken script, failing install, or instructions an agent cannot follow.
title: "[Bug]: "
labels: ["bug", "needs-triage"]
body:
  - type: markdown
    attributes:
      value: |
        Thanks for reporting this. Please do **not** use this form for security
        vulnerabilities — use [private vulnerability reporting](https://github.com/K-Dense-AI/scientific-agent-skills/security/advisories/new)
        instead, as described in [SECURITY.md](https://github.com/K-Dense-AI/scientific-agent-skills/blob/main/SECURITY.md).

  - type: dropdown
    id: area
    attributes:
      label: Area
      description: Which part of the repository is affected?
      options:
        - A skill under skills/
        - Repository tooling (scan_skills.py, tests, CI workflows)
        - Documentation (README, CONTRIBUTING, AGENTS)
        - Not sure
    validations:
      required: true

  - type: input
    id: skill
    attributes:
      label: Skill name
      description: The skill directory name, exactly as it appears under `skills/`. Leave blank if this is not skill-specific.
      placeholder: scanpy
    validations:
      required: false

  - type: textarea
    id: what-happened
    attributes:
      label: What happened
      description: What did the skill or tool actually do?
      placeholder: The skill's example call to sc.pp.neighbors() fails with a TypeError.
    validations:
      required: true

  - type: textarea
    id: expected
    attributes:
      label: What you expected instead
    validations:
      required: true

  - type: textarea
    id: reproduce
    attributes:
      label: Steps to reproduce
      description: The smallest sequence that triggers it. Include the prompt you gave the agent, if relevant.
      placeholder: |
        1. Load the `scanpy` skill in Claude Code
        2. Ask: "cluster my AnnData object"
        3. Run the code the agent produces
        4. See the error below
    validations:
      required: true

  - type: textarea
    id: logs
    attributes:
      label: Error output
      description: Paste the traceback or scanner output. This is rendered as a code block, so no backticks are needed.
      render: shell
    validations:
      required: false

  - type: textarea
    id: environment
    attributes:
      label: Environment
      description: Skill behavior varies by agent host and model, so please tell us where you saw this.
      value: |
        - Agent host (Claude Code, Cursor, Codex, other):
        - Model:
        - Repository version or commit:
        - Python version:
        - Operating system:
    validations:
      required: true

  - type: checkboxes
    id: checks
    attributes:
      label: Before submitting
      options:
        - label: I searched existing issues and this is not a duplicate.
          required: true
        - label: This is not a security vulnerability. (Those go through private reporting.)
          required: true