chore: change llm provider to groq

[felipefernandes]

No description provided.

[sentry]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[github-actions]

🧜‍♀️ Iara Code Review

Note: Inline mode failed (invalid JSON response from LLM)

{
  "summary": "❌ Review Failed: Found configuration changes and potential security issues",
  "comments": [
    {
      "file": ".github/workflows/iara-review.yml",
      "line": 20,
      "severity": "security",
      "is_blocking": true,
      "message": "🔒 Potential security issue: hardcoded secret reference in workflow file, consider using environment variables or secrets management"
    },
    {
      "file": ".github/workflows/iara-review.yml",
      "line": 21,
      "severity": "config",
      "is_blocking": false,
      "message": "💡 Configuration change: provider changed from 'anthropic' to 'groq', ensure compatibility and test thoroughly"
    },
    {
      "file": ".iara.json",
      "line": 27,
      "severity": "config",
      "is_blocking": false,
      "message": "💡 Configuration change: model provider changed from 'anthropic' to 'groq', ensure compatibility and test thoroughly"
    },
    {
      "file": ".iara.json",
      "line": 28,
      "severity": "config",
      "is_blocking": false,
      "message": "💡 Configuration change: preferred model changed from 'claude-sonnet-4-5' to 'llama-3.3-70b-versatile', ensure compatibility and test thoroughly"
    }
  ]
}

---

Reviewed by Iara - AI Code Reviewer

[github-actions]

🐛 Potential bug: The 'groq_api_key' is being used directly from secrets. Ensure that this key is properly secured and not exposed in the workflow file.

[github-actions]

✨ Code style improvement: The regular expression used to extract JSON from markdown blocks could be more efficient. Consider using a more specific pattern to avoid potential false positives.

[github-actions]

🧜‍♀️ Iara Code Review

❌ Found 2 issues: 1 potential bug and 1 style improvement

---

Reviewed by Iara - AI Code Reviewer

[github-actions]

🧜‍♀️ Iara Code Review

❌ Review Failed: Found issues with error handling and code style

---

.github/workflows/iara-review.yml:21
💡 Consider adding a comment explaining the reason for switching from 'anthropic' to 'groq' provider

.iara.json:27
💡 Consider adding a comment explaining the reason for switching from 'claude-sonnet-4-5' to 'llama-3.3-70b-versatile' model

iara/parsers/inline_parser.py:43
🐛 The new regex pattern may not handle all edge cases, consider adding more test cases to ensure its correctness

iara/parsers/inline_parser.py:104
🐛 The coercion of invalid severity to 'other' may mask real issues, consider logging the original severity value for debugging purposes

tests/test_inline_parser.py:143
✨ The test case for invalid severity value should be updated to reflect the new behavior of coercing to 'other'

---

Reviewed by Iara - AI Code Reviewer

[github-actions]

🧜‍♀️ Iara Code Review

❌ Found 2 issues: 1 potential security concern and 1 code quality issue

---

Reviewed by Iara - AI Code Reviewer

[github-actions]

🐛 Potential bug: The parse_inline_review function now coerces invalid severity values to 'other' instead of raising a ValueError. This might lead to unexpected behavior if the severity is not properly validated.

[github-actions]

🧜‍♀️ Iara Code Review

📝 Review of code diff: Found several changes, including updates to provider and API keys, modifications to inline review parsing, and test updates

---

.github/workflows/iara-review.yml:20
💡 Updated provider from 'anthropic' to 'groq' and API key from 'CLAUDE_API_KEY' to 'GROQ_API_KEY'

.iara.json:27
💡 Updated model provider from 'anthropic' to 'groq' and preferred model from 'claude-sonnet-4-5' to 'llama-3.3-70b-versatile'

iara/parsers/inline_parser.py:39
✨ Modified inline review parsing to extract JSON from markdown blocks or curly braces, and to coerce invalid severities to 'other'

tests/test_inline_parser.py:143
💡 Updated test for invalid severity value to check for coercion to 'other' instead of raising an exception

tests/test_inline_parser.py:187
💡 Added tests for parsing JSON embedded in markdown or conversational text, and for fallback extraction with curly braces

---

Reviewed by Iara - AI Code Reviewer