Agent building

Automated vulnerability scanner for agent platforms.

Cut a DaVinci Resolve talking-head / screencast / tutorial timeline into a clean, well-paced foundation edit - removing bad takes, retakes, false starts…

Measure and shrink the always-loaded context of a Claude Code harness (CLAUDE.md + rules without paths frontmatter) back under budget — migrate narrative to…

Session relay — compresses what's been done so far (writes a one-liner for /compact) and generates a "next task" prompt (a baton) for the user to hand off.

Analyze session work and automatically convert reusable patterns into Claude Code skills.

Drift detection + baseline integrity guard for agent workspace files with automatic alerting support

Captures learnings, errors, corrections, and feature requests to enable continuous improvement.

Automated sequencing quality control and advanced visualization wrapping FastQC, MultiQC, and custom chart generation.

Eval-driven skill tuning.

Odai 的可选多能力编排。用户要求分工、独立规划或审查、并行制作，或为 Codex、Claude Code、Copilot 配置模型路由时使用；治理技能 odai 独立可用。

Posts content and articles to X (Twitter).

Knowledge and governance closeout: reconcile project docs, rule files (CLAUDE.md/AGENTS.md), authorized agent memory, and workspace residue with what the code…

Selects bounded, graph-informed source slices with Trailmark and delegates focused code analysis or patch-proposal work to a smaller subagent.

Create, refactor, or debug agent skills and their loading.

Requested GitHub PR/issue agent transcripts: redact, trim, preview, and insert safely.

Summarize CodexBar local cost logs by model for Codex or Claude, including current or full breakdowns.

Author or review AgentSkills: create, repair, validate, or restructure SKILL.md files and bundled resources.

This skill should be used for diagnosing and mitigating context degradation: lost-in-middle failures, context poisoning, context clash, context confusion…

This skill should be used when building agent evaluation systems: deterministic checks, regression suites, multi-dimensional rubrics, quality gates…

Use when the user wants multi-agent division of labor for research-led work and the lead should stay on the critical path while 1-2 bounded sidecars handle…

Generate an image from a prompt or edit an existing one (Gemini image model, standard library only, any Gemini key the resolver finds) and attach the file to…

Bring the owner's Claude Code history into Sutando: index, extract and haiku-summarise the stock ~/.claude/projects transcripts, stage the result, show the…

Before answering your OWNER, when the answer genuinely depends on something another agent of the same owner holds and you do not, ask that one agent once…

Turn this agent into a shareable Claw for a person to review.