Feat(feature): Add sync workflow and script to automatically sync AgentKit templates to lamatic.ai agentkits#395
Conversation
…ntKit templates to lamatic.ai agentkits
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
WalkthroughAdds a new scheduled and on-demand GitHub Actions workflow plus a Python sync script that mirrors local JSON agent kit templates in Changes
Sequence Diagram(s)sequenceDiagram
participant GHA as GitHub Actions
participant Script as sync-agentkits.py
participant LocalFS as Local Filesystem
participant API as lamatic.ai API
GHA->>Script: Run (env: LAMATIC_API_KEY)
rect rgba(220,235,255,0.35)
note over Script,LocalFS: Discover local templates
Script->>LocalFS: Read `files/templates/*.json`
LocalFS-->>Script: File contents
Script->>Script: Compute SHA-256 per file
end
rect rgba(220,255,235,0.35)
note over Script,API: Fetch remote templates
Script->>API: GET /templates/agentkits (Authorization)
API-->>Script: Remote templates list
end
rect rgba(255,245,215,0.35)
note over Script,API: Reconcile and act
alt New local file
Script->>API: POST /templates/agentkits
else Updated local file
Script->>API: PUT /templates/agentkits/{id}
else Deleted local file
Script->>API: DELETE /templates/agentkits/{id}
end
Script-->>GHA: Log results & exit status
end
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes
Poem
Pre-merge checks and finishing touches❌ Failed checks (1 inconclusive)
✅ Passed checks (4 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 4
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
Disabled knowledge base sources:
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (2)
.github/workflows/sync-agentkits.yml(1 hunks)scripts/sync-agentkits.py(1 hunks)
🧰 Additional context used
🪛 Ruff (0.14.2)
scripts/sync-agentkits.py
41-41: Probable use of requests call without timeout
(S113)
48-48: Consider moving this statement to an else block
(TRY300)
50-50: Do not catch blind exception: Exception
(BLE001)
65-65: Probable use of requests call without timeout
(S113)
67-67: Do not catch blind exception: Exception
(BLE001)
74-74: Probable use of requests call without timeout
(S113)
76-76: Do not catch blind exception: Exception
(BLE001)
85-85: Probable use of requests call without timeout
(S113)
87-87: Do not catch blind exception: Exception
(BLE001)
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (3)
.github/workflows/sync-agentkits.yml (1)
27-30: Consider caching pip dependencies.Adding pip cache can speed up workflow runs, especially for the scheduled daily sync.
Apply this diff to add caching:
- name: Set up Python uses: actions/setup-python@v5 with: python-version: '3.10' + cache: 'pip' + cache-dependency-path: '**/requirements.txt' - name: Install dependencies run: | python -m pip install --upgrade pip pip install requestsAlternatively, if you create a
requirements.txtfile withrequestslisted, the cache will work automatically.scripts/sync-agentkits.py (2)
92-100: Consider safeguard for bulk deletions.The deletion logic correctly removes remote templates that no longer exist locally, but a misconfiguration (wrong directory, empty checkout, filesystem issues) could trigger mass deletion of production templates.
Add a safeguard to abort if too many deletions are attempted:
+ # Safeguard: abort if more than 50% of templates would be deleted + deletion_count = len([f for f in remote_templates if f not in local_files]) + if deletion_count > 0 and deletion_count >= len(remote_templates) * 0.5: + print(f"ABORT: Would delete {deletion_count}/{len(remote_templates)} templates (>50%). " + "This may indicate a misconfiguration.") + return False + for filename in remote_templates: if filename not in local_files: print(f"DELETING: {filename}")Alternatively, add a
--no-deleteflag or require an environment variable (e.g.,ALLOW_DELETIONS=true) before performing any deletions.
70-78: Enhance error messages with HTTP details.The current error messages don't include HTTP status codes or response bodies, making it harder to diagnose API failures during sync runs.
Apply this diff to add detailed error information:
if filename not in remote_templates: print(f"CREATING: {filename}") try: payload = {"name": filename, "content": content, "hash": local_data['hash']} response = requests.post(API_BASE_URL, headers=HEADERS, json=payload, timeout=REQUEST_TIMEOUT) response.raise_for_status() except Exception as e: - print(f"Error creating {filename}: {e}") + status = getattr(getattr(e, 'response', None), 'status_code', 'N/A') + print(f"Error creating {filename}: {e} (HTTP {status})") sync_failed = True elif local_data['hash'] != remote_templates[filename]: print(f"UPDATING: {filename}") try: payload = {"content": content, "hash": local_data['hash']} response = requests.put(f"{API_BASE_URL}/{filename}", headers=HEADERS, json=payload, timeout=REQUEST_TIMEOUT) response.raise_for_status() except Exception as e: - print(f"Error updating {filename}: {e}") + status = getattr(getattr(e, 'response', None), 'status_code', 'N/A') + print(f"Error updating {filename}: {e} (HTTP {status})") sync_failed = True ... try: response = requests.delete(f"{API_BASE_URL}/{filename}", headers=HEADERS, timeout=REQUEST_TIMEOUT) response.raise_for_status() except Exception as e: - print(f"Error deleting {filename}: {e}") + status = getattr(getattr(e, 'response', None), 'status_code', 'N/A') + print(f"Error deleting {filename}: {e} (HTTP {status})") sync_failed = TrueAlso applies to: 80-88, 92-100
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
Disabled knowledge base sources:
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (2)
.github/workflows/sync-agentkits.yml(1 hunks)scripts/sync-agentkits.py(1 hunks)
🧰 Additional context used
🧬 Code graph analysis (1)
scripts/sync-agentkits.py (1)
lib/templateData.ts (1)
templates(16-185)
🪛 Ruff (0.14.2)
scripts/sync-agentkits.py
50-50: Consider moving this statement to an else block
(TRY300)
52-52: Do not catch blind exception: Exception
(BLE001)
65-65: Do not catch blind exception: Exception
(BLE001)
76-76: Do not catch blind exception: Exception
(BLE001)
86-86: Do not catch blind exception: Exception
(BLE001)
98-98: Do not catch blind exception: Exception
(BLE001)
🔇 Additional comments (2)
scripts/sync-agentkits.py (2)
10-10: LGTM!The timeout constant has been added as suggested in previous reviews, preventing indefinite hangs.
59-59: LGTM!The failure tracking and non-zero exit code have been added as suggested in previous reviews. The script now correctly reports sync failures to CI.
Also applies to: 102-102, 108-112
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (4)
requirements.txt (1)
1-1: Consider pinning the requests version.Unpinned dependencies can lead to non-reproducible builds and potential breakage from future releases. Since this is a production automation script, pinning to a specific version (or range) improves reliability.
Apply this diff:
-requests +requests>=2.32.0,<3.0.0Alternatively, pin to an exact version if strict reproducibility is required:
-requests +requests==2.32.3.github/workflows/sync-agentkits.yml (2)
1-15: Add concurrency control to prevent simultaneous executions.The workflow lacks concurrency control, so a manual
workflow_dispatchtrigger could run simultaneously with a scheduled execution or a push-triggered run. Concurrent syncs may cause race conditions or conflicting API operations (e.g., both attempting to update the same template).Apply this diff to add concurrency control:
name: Sync AgentKit Templates +concurrency: + group: sync-agentkit-templates + cancel-in-progress: false + on: schedule:This ensures only one sync runs at a time, queuing subsequent triggers until the current run completes.
30-39: LGTM!Dependency installation and script execution are correctly configured. The API key is properly sourced from repository secrets, and the script path is accurate.
The inline comment on line 33 (
# === FIX 2: Install from requirements.txt ===) appears to be a development artifact. Consider removing it for cleaner production code:- # === FIX 2: Install from requirements.txt === pip install -r requirements.txtscripts/sync-agentkits.py (1)
86-92: Consider removing redundant Exception handlers.After catching
RequestException(which covers most requests-related failures), the subsequentexcept Exceptionblocks (lines 90-92, 104-106, 126-128) rarely catch additional errors. The pattern is safe but verbose.If you prefer to keep broad exception catching, the current code is acceptable. For slightly cleaner code, you could consolidate:
except requests.exceptions.RequestException as e: status = getattr(getattr(e, 'response', None), 'status_code', 'N/A') print(f"Error creating {rel_path}: {e} (HTTP {status})") sync_failed = True - except Exception as e: - print(f"Error creating {rel_path}: {e}") - sync_failed = TrueApply the same pattern to lines 104-106 and 126-128.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
Disabled knowledge base sources:
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (3)
.github/workflows/sync-agentkits.yml(1 hunks)requirements.txt(1 hunks)scripts/sync-agentkits.py(1 hunks)
🧰 Additional context used
🧬 Code graph analysis (1)
scripts/sync-agentkits.py (1)
lib/templateData.ts (1)
templates(16-185)
🪛 Ruff (0.14.2)
scripts/sync-agentkits.py
56-56: Consider moving this statement to an else block
(TRY300)
62-62: Do not catch blind exception: Exception
(BLE001)
75-75: Do not catch blind exception: Exception
(BLE001)
90-90: Do not catch blind exception: Exception
(BLE001)
104-104: Do not catch blind exception: Exception
(BLE001)
126-126: Do not catch blind exception: Exception
(BLE001)
🔇 Additional comments (7)
.github/workflows/sync-agentkits.yml (2)
16-21: LGTM!Job configuration and checkout step are correctly configured.
23-28: LGTM!Python setup with caching is correctly configured and will optimize workflow execution time.
scripts/sync-agentkits.py (5)
1-19: LGTM!The imports, constants, and API key validation are well-structured. The early exit on missing credentials prevents silent failures and provides clear error messaging.
21-24: LGTM!The hash calculation is correctly implemented using SHA-256 on binary file content, ensuring consistent hashing across platforms.
26-43: LGTM!The function correctly handles subdirectories and normalizes paths with forward slashes for cross-platform consistency. The directory existence check prevents cryptic errors downstream.
110-114: Excellent safeguard against accidental mass deletion!This protection prevents catastrophic data loss if the local directory is empty or misconfigured. The 50% threshold and clear messaging make this a robust safety mechanism.
132-140: LGTM!The main block orchestrates the sync cleanly with appropriate exit codes for CI/CD integration. The error handling ensures workflow failures are properly reported to GitHub Actions.
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (2)
scripts/sync-agentkits.py (2)
28-44: Consider validating JSON files during discovery.The function successfully addresses the previous filename collision issue by using relative paths. However, it doesn't validate that discovered JSON files are well-formed or contain expected fields. A corrupted JSON file will only fail later during sync, making debugging harder.
Consider adding basic validation:
for root, _, files in os.walk(LOCAL_TEMPLATES_DIR): for filename in files: if filename.endswith(".json"): filepath = os.path.join(root, filename) - file_hash = get_file_hash(filepath) - rel_path = os.path.relpath(filepath, LOCAL_TEMPLATES_DIR).replace(os.sep, '/') - local_files[rel_path] = {"path": filepath, "hash": file_hash} + try: + # Validate JSON is well-formed + with open(filepath, 'r') as f: + json.load(f) + file_hash = get_file_hash(filepath) + rel_path = os.path.relpath(filepath, LOCAL_TEMPLATES_DIR).replace(os.sep, '/') + local_files[rel_path] = {"path": filepath, "hash": file_hash} + except (json.JSONDecodeError, OSError) as e: + print(f"Warning: Skipping invalid file {filepath}: {e}") + continue
112-128: Excellent safeguard against accidental mass deletion.The 50% threshold protection is a smart defensive measure that prevents catastrophic mistakes from configuration errors or bugs. The deletion logic properly encodes paths and handles errors consistently.
Consider extracting the threshold as a named constant for clarity:
+# Safeguard: abort if attempting to delete more than this percentage +MAX_DELETION_THRESHOLD = 0.5 + def sync_templates(local_files, remote_templates): """Compares local and remote and performs sync operations.""" ... deletion_count = len([path for path in remote_templates if path not in local_files]) - if deletion_count > 0 and deletion_count >= (len(remote_templates) * 0.5): + if deletion_count > 0 and deletion_count >= (len(remote_templates) * MAX_DELETION_THRESHOLD): print(f"ABORT: Safeguard triggered. Attempting to delete {deletion_count}/{len(remote_templates)} templates (>50%).")
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
Disabled knowledge base sources:
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (3)
.github/workflows/sync-agentkits.yml(1 hunks)requirements.txt(1 hunks)scripts/sync-agentkits.py(1 hunks)
🚧 Files skipped from review as they are similar to previous changes (2)
- requirements.txt
- .github/workflows/sync-agentkits.yml
🧰 Additional context used
🧬 Code graph analysis (1)
scripts/sync-agentkits.py (1)
lib/templateData.ts (1)
templates(16-185)
🪛 Ruff (0.14.2)
scripts/sync-agentkits.py
62-62: Consider moving this statement to an else block
(TRY300)
82-82: Do not catch blind exception: Exception
(BLE001)
🔇 Additional comments (6)
scripts/sync-agentkits.py (6)
23-26: LGTM!The hash calculation is straightforward and correct. Opening in binary mode ensures consistent hashing across platforms.
46-71: Solid error handling with appropriate fail-fast behavior.The function properly validates content-type before attempting JSON parsing, addressing the previous HTML response concern. Exception handling distinguishes between network errors and JSON decoding failures with clear error messages.
Minor note: Line 60 assumes each template object has
nameandhashfields. If the API returns objects with different keys, this will raise aKeyError.
87-96: LGTM!The create operation properly handles errors and includes the timeout. The defensive
getattrpattern for extracting the status code prevents exceptions when the response is None.
98-108: LGTM!The update operation correctly encodes the path while preserving forward slashes (
safe='/'), which is essential for subdirectory support. Error handling is consistent and comprehensive.
132-140: LGTM!The main block properly orchestrates the sync workflow and exits with the correct status code for CI/CD integration. Clear status messages aid troubleshooting.
8-11: Let me try an unauthenticated request to at least confirm the endpoint is reachable:Let me search the repository for more context about this script's intended functionality:
The API endpoint may not exist or may be incorrectly designed.
Lamatic.ai's documented API exclusively uses GraphQL integration, but the script uses REST-style patterns (
https://lamatic.ai/templates/agentkits) that don't appear in any official documentation. The script cannot be verified without valid credentials (which are not available in the sandbox environment).Before deploying, manually verify that:
- The endpoint actually exists and returns JSON (not HTML)
- The GraphQL API should be used instead if this is the preferred Lamatic integration pattern
- Previous reviews flagged this endpoint as returning HTML—confirm this has been resolved with the service owner
There was a problem hiding this comment.
Actionable comments posted: 1
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
Disabled knowledge base sources:
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (1)
scripts/sync-agentkits.py(1 hunks)
🧰 Additional context used
🧬 Code graph analysis (1)
scripts/sync-agentkits.py (1)
lib/templateData.ts (1)
templates(16-185)
🪛 Ruff (0.14.2)
scripts/sync-agentkits.py
62-62: Consider moving this statement to an else block
(TRY300)
🔇 Additional comments (5)
scripts/sync-agentkits.py (5)
8-21: Configuration looks solid.The API key validation and header setup are appropriate. The REQUEST_TIMEOUT constant and defensive Content-Type checking (handled later in the code) address past concerns about API reliability.
23-26: LGTM!Clean hash implementation using SHA-256.
28-44: Well implemented with proper safeguards.The directory validation and relative path handling correctly address the filename collision concern from earlier reviews. Path normalization ensures cross-platform compatibility.
73-130: Excellent implementation with outstanding defensive measures.The sync logic correctly handles all CRUD operations with proper error tracking. The mass-deletion safeguard on lines 112-116 is exemplary defensive programming—aborting when >50% of templates would be deleted prevents catastrophic data loss from misconfigurations.
The specific exception handling (line 82), URL encoding (lines 102, 122), and detailed HTTP status logging all demonstrate production-ready quality.
132-140: LGTM!Clean orchestration with appropriate exit codes. The conditional exit on line 140 ensures CI will correctly detect sync failures.
There was a problem hiding this comment.
Actionable comments posted: 1
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
Disabled knowledge base sources:
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (1)
scripts/sync-agentkits.py(1 hunks)
🧰 Additional context used
🧬 Code graph analysis (1)
scripts/sync-agentkits.py (1)
lib/templateData.ts (1)
templates(16-185)
🪛 Ruff (0.14.2)
scripts/sync-agentkits.py
68-68: Consider moving this statement to an else block
(TRY300)
| deletion_count = len([path for path in remote_templates if path not in local_files]) | ||
| if deletion_count > 0 and deletion_count >= (len(remote_templates) * 0.5): | ||
| print(f"ABORT: Safeguard triggered. Attempting to delete {deletion_count}/{len(remote_templates)} templates (>50%).") | ||
| print("This may indicate a misconfiguration. Aborting all deletions.") | ||
| sync_failed = True | ||
| else: | ||
| for rel_path in remote_templates: | ||
| if rel_path not in local_files: | ||
| print(f"DELETING: {rel_path}") | ||
| try: | ||
| encoded_path = quote(rel_path, safe='/') | ||
| response = requests.delete(f"{API_BASE_URL}/{encoded_path}", headers=HEADERS, timeout=REQUEST_TIMEOUT) | ||
| response.raise_for_status() | ||
| except requests.exceptions.RequestException as e: | ||
| status = getattr(getattr(e, 'response', None), 'status_code', 'N/A') | ||
| print(f"Error deleting {rel_path}: {e} (HTTP {status})") | ||
| sync_failed = True |
There was a problem hiding this comment.
Deletion safeguard blocks valid deletions
With the current >50% safeguard, a perfectly legitimate sync is prevented—for example, if the remote set has 1 template and we remove it locally, deletion_count becomes 1, which is ≥ 50% of 1, so the script aborts and exits 1. Any restructuring that removes a majority of templates will fail the same way, violating the requirement that deletions mirror the repo. Please loosen the guard so it only trips when both the percentage and an absolute count threshold are exceeded.
- if deletion_count > 0 and deletion_count >= (len(remote_templates) * 0.5):
+ if (
+ deletion_count > 0
+ and len(remote_templates) >= 10
+ and deletion_count >= (len(remote_templates) * 0.5)
+ ):1 participant
Fixes #376
Description
This PR introduces a complete GitHub Actions workflow to automatically synchronize the agentkit templates from the /files/templates directory to the https://lamatic.ai/templates/agentkits endpoint, as requested in issue #376.
Changes Implemented
Two New Files
This is my contribution for Hacktoberfest 2025.
Summary by CodeRabbit