菜单
Materialize the local video-evaluator skill pack into another repo so Codex or Claude Code can use the shared review skills there.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Assemble a grounded review prompt from a local artifact bundle so Codex or Claude Code can inspect the right files and ask the right questions.
Check a local video against layout annotations, safe zones, OCR text boxes, and frame samples so agents can catch cropped content, caption overlap, unreadable overlays, and platform-unsafe composition before shipping.
Collect first-pass source media facts, audio silence/energy, shot estimates, representative frame status, and persistent text/caption risk evidence.
Review a rendered video for technical defects such as resolution mismatch, missing or near-silent audio, black or white frames, edge gutters, low motion, sparse caption bands, and layout issue pass-through.
Fuse video shot segments with storyboard frames, OCR, transitions, and timeline artifacts into one per-segment evidence map for grounded video review.
Extract at least one storyboard frame per video shot segment so later OCR and segment evidence fusion can cover gaps missed by global sampling.
| name | install-skill-pack |
| description | Materialize the local video-evaluator skill pack into another repo so Codex or Claude Code can use the shared review skills there. |
Use this when another repo should consume the local skill pack instead of re-implementing review helpers.
Outputs:
skills/*dist/*agent/run-tool.mjsRequired:
{
"targetDir": ".video-evaluator",
"includeAgentRunner": true,
"installDependencies": true
}
targetDir is relative to the current working directory unless absolute.
includeAgentRunner defaults to true. installDependencies defaults to
true.
The result lists copied files, including skill docs/examples, runtime files, and
the optional agent runner. When dependency installation is enabled, the target
pack should be runnable through node ./.video-evaluator/agent/run-tool.mjs.
Run this from the source video-evaluator repo when another repo needs the
pack. After installing, switch to the target repo and run skill-catalog via
the installed-pack runner to verify discovery. Then run the narrower skill
needed for the target repo's artifacts.
This is an installation/materialization helper, not a review skill. It copies the current local skill docs and built runtime, so source changes that have not been built may not be reflected in copied runtime behavior.
Do not run this inside a target repo unless you intend to modify or create the
pack directory there. Do not install over another agent's in-progress changes
without checking git status in the target repo.
Missing built dist files, permission errors, or package manager failures can
leave a partial install. If dependency installation is disabled, the target repo
must already have compatible runtime dependencies. Re-run with an explicit
targetDir if the pack lands in the wrong location.
Repo-side runner:
node --import tsx scripts/harness/install-skill-pack.ts
Installed-pack runner:
node ./.video-evaluator/agent/run-tool.mjs install-skill-pack