6 results for “evals”
Evaluate, score, and systematically improve prompts in the codebase. Identifies weak prompts, generates test cases, scores outputs, and proposes optimized versions. Use when the user says "improve this prompt", "why is the AI doing X", "eval my prompts", or "optimize the agent".
Write publishable content for a Senso prompt from the knowledge base, and check it before it goes anywhere. The agent drafts from retrieved sources — in the shape of the content type and the voice of the brand kit — states nothing the knowledge base does not back, saves the draft to Senso and sends the user a link to read it rather than printing it into the conversation, and evaluates it once after the user has reviewed it. Use for drafting content for a tracked question ("write something for this prompt", "draft an article", "generate content"), for revising a draft, and for judging a piece of text against the knowledge base or the brand kit ("is this accurate", "check this text against our docs", "run an eval"). Also owns the content type and brand kit commands, which senso-verification-loop-setup runs as chores and this skill runs when a draft is asked for on its own. The drafting and evaluation step of senso-verification-loop. Not for publishing — that is senso-publish.
Retired. This skill's contents moved to three skills — senso-verification-loop-setup (the industry, the brand name, the segment), senso-verification-loop (where the brand stands, the next question it is missing from, the accuracy of answers that name it) and senso-gap-report (the gap report). It no longer does anything. If it is loaded, tell the user to run `senso setup` to install the current skills, and use those.
Texas Hold'em poker intelligence — hand evaluation, pot odds, position strategy, and live player profiling for AI agents
Guides agents through autonomous ManiSkill and VSLAM evaluation, tuning, verification, memory storage, and summary writing using the Autolab MCP server and repo tooling.
Iteratively optimize thermal designs by solving 2D heat equations (Poisson PDE). Parameterize heat source placement and material conductivity, simulate temperature distributions, evaluate performance metrics, and propose improvements autonomously. Use when the user asks to design, optimize, or analyze heat sinks, thermal layouts, cooling systems, or any steady-state thermal problem on a 2D domain.