Add ollama-action GitHub Actions workflow for LLM automation

[Copilot]

Adds a workflow using ai-action/ollama-action@v2 to run Ollama LLMs in CI — useful for summarizing PRs, generating changelogs, or other AI-assisted automation tasks.

Workflow (.github/workflows/ollama-action.yml)

• Triggers: workflow_dispatch (manual, with configurable inputs) + push to main
• Inputs: model, prompt, version, cache — all optional with sensible defaults (llama3.2, model caching enabled)
• Steps: checkout → run LLM via composite action → print response → save to .github/ollama-response.txt
• Permissions: contents: read only (least privilege — file is workspace-only, not committed back)
• Output handling uses env vars (not direct ${{ }} in run:) to avoid script injection

.gitignore

• Added .github/ollama-response.txt to prevent accidental commits of generated output

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Add GitHub Action workflow: ollama-action for LLM automation</issue_title>
<issue_description>## Proposal
Add an ollama-action GitHub Action workflow to the repository. This workflow will allow maintainers and contributors to run Ollama large language models (LLMs) on repo events or manually using the ai-action/ollama-action composite action.

Rationale

• Enables automation/AI tasks directly within the CI pipeline (e.g. summarize PRs, changelogs, generate content).
• Simple, secure, and fast setup without maintaining your own LLM infrastructure (uses latest Ollama runner on GitHub Actions).
• The action supports model caching, customizable prompts, and Ollama version selection.

Proposed Workflow Features

• Manual trigger (workflow_dispatch) and optional automatic run on push to main
• Inputs for model name, prompt, version, and cache toggle
• Prints LLM output to workflow logs and saves to .github/ollama-response.txt for downstream use

Implementation Example

name: ollama-action
on:
  workflow_dispatch:
    inputs:
      model:
        description: 'Ollama model name (e.g. llama3.2, tinyllama)'
        required: false
        default: 'llama3.2'
      prompt:
        description: 'Prompt to run'
        required: false
        default: 'Summarize recent changes in this repository.'
      version:
        description: 'Optional Ollama version to install (e.g. 0.15.0)'
        required: false
        default: ''
      cache:
        description: 'Whether to cache the model (true/false)'
        required: false
        default: 'true'
  push:
    branches:
      - main
jobs:
  run-ollama:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Run Ollama (ai-action/ollama-action)
        id: ollama
        uses: ai-action/ollama-action@v2
        with:
          model: ${{ github.event.inputs.model || 'llama3.2' }}
          prompt: ${{ github.event.inputs.prompt || 'Summarize recent changes in this repository.' }}
          version: ${{ github.event.inputs.version }}
          cache: ${{ github.event.inputs.cache || 'true' }}
      - name: Print Ollama response
        run: |
          echo "=== Ollama response ==="
          echo "${{ steps.ollama.outputs.response }}"
      - name: Save response to file
        run: |
          mkdir -p .github
          printf "%s\n" "${{ steps.ollama.outputs.response }}" > .github/ollama-response.txt

Reference: ai-action/ollama-action action files

• action.yml: defines inputs (model, prompt, version, cache), output (response), and steps to setup Ollama, cache models, run the selected LLM, and output LLM results to workflow output.
• README.md: provides examples for workflow usage, customization, and explains required/optional parameters.
• Additional config: Also reviewed project configs under .github/ such as templates, dependabot and mergify, all standard.

---

See ai-action/ollama-action for upstream details and ongoing updates.

</issue_description>

<agent_instructions>Review the whole repo codebase and perform online research for ollama integrations along with authentication (if needed) and the whole repository https://github.com/ai-action/ollama-action/tree/master then start implementing the task </agent_instructions>

Comments on the Issue (you are @copilot in this section)

• Fixes Add GitHub Action workflow: ollama-action for LLM automation #358

---

📍 Connect Copilot coding agent with Jira, Azure Boards or Linear to delegate work to Copilot in one click without leaving your project management tool.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
stormcomui	Error		Mar 14, 2026 10:53pm

[Copilot]

Pull request overview

Adds a GitHub Actions workflow to run Ollama LLMs in CI (manual dispatch and on pushes to main), capturing the model response for downstream automation use.

Changes:

• Added .github/workflows/ollama-action.yml workflow using ai-action/ollama-action@v2 with configurable inputs.
• Persisted the LLM response to .github/ollama-response.txt and printed it to logs.
• Ignored the generated response file via .gitignore to prevent accidental commits.

Reviewed changes

Copilot reviewed 1 out of 2 changed files in this pull request and generated 2 comments.

File	Description
.gitignore	Ignores the generated Ollama response output file.
.github/workflows/ollama-action.yml	Introduces the Ollama CI workflow with inputs, execution, logging, and response persistence.

---

You can also share your feedback on Copilot code review. Take the survey.

[syed-reza98]

@copilot apply changes based on the comments in this thread

[Copilot]

@copilot apply changes based on the comments in this thread

Applied both review suggestions in e6a030f:

• Added timeout-minutes: 15 at job level to prevent stuck runs from consuming runner minutes indefinitely.
• Replaced echo with printf for robust output handling (safe against -n/-e flags), and added sed 's/^::/\x200::/' to sanitize lines starting with :: to prevent GitHub Actions workflow-command injection via log output.