Nm Imbue Feature Review

Review and prioritize features using RICE, WSJF, or Kano scoring frameworks, then create GitHub issues for suggestions

This skill appears to do what it says (inventory, score, suggest, optionally create GitHub issues), but verify the following before installing or running it with real repositories: 1) Authentication: Ask the author/maintainer where GitHub authentication comes from. If the skill will create issues you should ensure a dedicated token is used and that the token's scope is limited (repo:issues or fine-grained app token), rather than granting full repo/admin rights. The skill's metadata does not declare a GH token env var — confirm whether credentials are expected in project config paths (e.g., night-market.*) or provided by the agent platform. 2) Config & Secrets: Inspect any .feature-review.yaml or the referenced config paths (night-market.imbue:scope-guard, night-market.imbue:review-core, night-market.tome:research) in your project to ensure they do not contain sensitive credentials you don't want the skill to read. If those configs live in a shared location, confirm access controls. 3) Repo Access: The skill will scan code, docs, and git history. Run it first on a test/sandbox repo to observe behavior and outputs before pointing it at production repos. 4) Tome Integration: If you use the --research flag the skill will try to call the tome plugin via Skill() interfaces; confirm that tome is installed and that its permissions/data access are acceptable. 5) Autonomy: Because the skill can be invoked autonomously by agents, consider disabling autonomous invocation or restricting when/which agents can call it until you confirm credentials and behavior. If the maintainer can confirm where authentication is obtained and that tokens/configs are scoped and separated from sensitive material, the remaining concerns are minor.

Type: OpenClaw Skill Name: nm-imbue-feature-review Version: 1.0.0 The skill bundle provides a comprehensive framework for feature prioritization and roadmap management using industry-standard models like RICE, WSJF, and Kano. It includes detailed documentation for feature discovery, classification, and tradeoff analysis (modules/tradeoff-dimensions.md), and integrates with GitHub for issue creation with explicit user confirmation (SKILL.md). While it mentions automated execution of a local script (scripts/deferred_capture.py) for backlog management and optional external research via the tome plugin, these capabilities are well-documented and align strictly with the stated purpose of the tool without evidence of malicious intent or unauthorized data exfiltration.