skills-pager

Build or reuse a compact navigation index for large, multi-file, or layered skills so you can load only the sections you need instead of rereading the full s...

This skill appears coherent and low-risk for its stated purpose, but before installing or enabling it consider: 1) The agent (or you when running the scaffold script) will read target SKILL.md and any referenced files and will write .skill-index/registry.json and .skill-index/skills/<skill-id>/index.md into the workspace root — make sure you want those persistent files created. 2) The included script requires a Node runtime and filesystem write permission; you can run it manually to verify output before letting the agent run it. 3) Because the agent will read files it can access, ensure no sensitive secrets are stored in files you do not want to be scanned/indexed. 4) Review generated index.md and registry.json before relying on them; the scaffold creates placeholders that the agent/user should fill in. If you want stricter control, run the script yourself and inspect outputs rather than granting autonomous runs.

Type: OpenClaw Skill Name: skills-pager Version: 1.0.0 The skills-pager bundle is a productivity utility designed to help AI agents manage large skill files by creating a navigation index in a `.skill-index/` directory. The included Node.js script (`scripts/create-skills-pager-map.js`) performs standard file system operations to scaffold these indices and lacks any network activity, obfuscation, or unauthorized data access. The instructions in `SKILL.md` and the `references/` directory are strictly aligned with the stated purpose of context optimization and include explicit safety boundaries to prevent the agent from indexing sensitive files like session state or memory.