No. Skills (SKILL.md) teach the agent how to perform a task or workflow. Trustabl Agent Analyzer (TAA) makes the tools themselves reliable, safe, and production-ready. Skills focus on process. TAA focuses on resilience, validation, policy, observability, and supply-chain trust. The two are highly complementary — great skills need hardened tools underneath them.
First run trustabl scan . --format json > scan.json. Then run trustabl enrich --input scan.json --repo . --diff --apply with your own Anthropic, OpenAI, or Google key. Trustabl displays the proposed patch, confirms that the current file still matches the scanned version, and writes a .trustabl.bak backup before applying a fix. If the file changed after the scan, Trustabl skips the stale fix instead of applying it.
Trustabl’s open-source Agent Reliability Analyzer scans AI agent repositories for reliability, safety, configuration, and tool-calling issues before production. Scanning runs locally with no account, code upload, network connection, or LLM. To generate code fixes, run trustabl enrich --input scan.json --repo . --diff --apply with your own Anthropic, OpenAI, or Google key; Trustabl shows the patch before applying the approved changes.
Yes. OpenShell and Trustabl address different layers. OpenShell helps isolate the runtime; Trustabl’s analyzer checks agent code and configuration for reliability and safety issues before deployment. Automated OpenShell policy generation is planned for Trustabl Console, not part of the open-source scanner today.
