Automate retry handling for flaky tools in automated workflows.
npx skills add https://github.com/keshrisohit/omniforge --skill flaky-skill
Please help me install this Agent Skill. Skill: flaky-skill Source: https://github.com/keshrisohit/omniforge/tree/main/tests/skills/fixtures/test_skills/flaky-skill Command: npx skills add https://github.com/keshrisohit/omniforge --skill flaky-skill
This skill provides automated retry handling to cope with flaky tools in automated workflows, reducing manual intervention when transient failures occur.
Provide your task arguments and let the flaky-skill retry automatically until success.
High-intent search queries and answers about installing and using this skill.
Automated retry handling for flaky tools re-runs failed tasks up to configured max-iterations and per-tool max-retries limits. This reduces manual intervention when transient failures occur during autonomous task execution.
Yes, you can configure max retries per tool in CI/CD pipelines. The retry behavior honors configurable resilience limits including max-iterations and max-retries-per-tool to control how many times failed tool calls re-execute.
Automated retry logic for unstable services is a deterministic mechanism that re-runs task executions when intermittent failures happen. It applies to autonomous scenarios where tools fail transiently and require multiple iterations to succeed.
Retry handling works with data pipelines using unstable tools. It is designed for CI/CD steps and data pipelines relying on unstable services, automatically re-running tasks until success or configured retry limits are reached.
The limitation of automated retry behavior is that execution stops after reaching configured max-iterations and per-tool max-retries. It only addresses transient failures and cannot resolve persistent tool errors or execution flaws.
masharrattLimit Node.js heap size dynamically across CFN Loop tasks and tools.
NVIDIAAnalyze EAGLE3 pipeline logs to classify failures and summarize task outcomes.
act70255Automate cloud infrastructure, CI/CD pipelines, Kubernetes, monitoring, security, and cost optimization.
claudcherejiOrchestrate distributed error handling with circuit breaker, retry, and fallback mechanisms.
