Nm Tome Research

Multi-source research across code, discourse, and academic channels

This skill's workflow fits its description, but the SKILL.md assumes you have a Python runtime, a local 'tome' package providing many modules, and an 'Agent' dispatch tool — none of which are declared in the metadata. Before installing/using: (1) confirm your agent/runtime provides the 'tome' package and the Agent tool or add an explicit install step; (2) be aware the skill will write session files to your current working directory (docs/research/...), so run it in an appropriate or isolated workspace; (3) if you don't want automatic filesystem writes, request a dry-run or adjust the save path; (4) no credentials are requested by the skill, but lack of declared dependencies may cause silent failures — verify prerequisites first. I have medium confidence because the issues look like omissions/assumptions rather than clearly malicious intent; providing the runtime dependency list or an install spec would raise confidence to benign.

Type: OpenClaw Skill Name: nm-tome-research Version: 1.0.0 The skill is a research orchestrator designed to manage a multi-step workflow including domain classification, research planning, and parallel agent dispatching using the 'tome' library. The instructions in SKILL.md define a legitimate process for synthesizing findings from various sources (code, discourse, academic) and saving formatted reports to the local filesystem (docs/research/). No evidence of malicious intent, data exfiltration, or prompt injection was found in the provided files.