Ln 652 Transaction Correctness Auditor
This AI audits transactions for errors and inconsistencies, ensuring data accuracy and preventing costly mistakes.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add ln-652-transaction-correctness-auditor npx -- -y @trustedskills/ln-652-transaction-correctness-auditor
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"ln-652-transaction-correctness-auditor": {
"command": "npx",
"args": [
"-y",
"@trustedskills/ln-652-transaction-correctness-auditor"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
This skill, the Ln 652 Transaction Correctness Auditor, is a specialized AI worker designed to audit database transaction patterns for correctness and consistency. It checks commit patterns, transaction boundaries, rollback handling, and trigger/notify semantics within a codebase. The auditor produces structured findings detailing severity, location, effort required for remediation, and recommendations, alongside calculating a compliance score (out of 10) for the "Transaction Correctness" category.
When to use it
- When performing code reviews for database-driven applications.
- To proactively identify potential errors or inconsistencies in transaction logic before deployment.
- During security audits to ensure proper handling of database transactions and triggers.
Key capabilities
- Audits transaction correctness, scope, and trigger interaction.
- Checks commit patterns, transaction boundaries, rollback handling, and trigger/notify semantics.
- Generates structured findings with severity levels, location details, effort estimates, and remediation recommendations.
- Calculates a compliance score (X/10) for Transaction Correctness.
- Supports domain-aware analysis using
domain_modeandcurrent_domain. - Requires loading
shared/references/audit_worker_core_contract.mdandshared/references/two_layer_detection.md.
Example prompts
- "Audit the transaction logic in this codebase for correctness."
- "Check the commit patterns and rollback handling within these migration files."
- "Analyze trigger-affected tables and their associated UPDATE paths."
Tips & gotchas
- Mandatory Read: The skill requires loading
shared/references/audit_worker_core_contract.mdandshared/references/two_layer_detection.mdto function correctly. Failure to load these will result in inaccurate or incomplete results. - The skill is a Level 3 (L3) worker, indicating it's designed for more specialized tasks.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.