All skills
agent-observability
Adds local hook-based metrics for tool latency, token usage, session cost and multi-agent traces.
- Part of Tencent/LoopForge
- AI Engineering · Evals and monitoring
More Evals and monitoring skills
See all- skillevaluator287Runs SkillEvaluator to score a SKILL.md, write test cases, explain reports and gate skills in CI.
- writing-instructions19Rewrites prompts and instructions so the model actually follows them.
- create-custom-grader8Converts an existing benchmark, rubric, verifier or task file into a SkillEvaluator custom evaluation.
- evolving-with-swarms1Improves a program against your scorer by running cheap model swarms within a dollar cap.
Also in this pack
- tech-design3Writes a complete technical design document in Markdown from a requirement report or verbal description.
- report-workflow-usage1Reports real model usage from finished external workflows to AgentFlow, including retries and multi-model totals.
- brainstormingExplores your intent, requirements and design with you before any feature or change is built.
- devflowRuns a structured, resumable delivery process from requirements and design through review, testing and preview deploy.