rebrand-sweep

Scan repositories for legacy brand residue and generate categorized findings reports.

Updated Apr 11, 2026
One-click install
npx skills add https://github.com/gozonerd/shadow-ai-assessment --skill rebrand-sweep-gozonerd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rebrand-sweep
Source: https://github.com/gozonerd/shadow-ai-assessment/tree/main/.claude/skills/rebrand-sweep
Command: npx skills add https://github.com/gozonerd/shadow-ai-assessment --skill rebrand-sweep-gozonerd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Forking or rebranding a codebase leaves behind legacy brand names in filenames, frontmatter, manifests, CI configs, binary metadata, and git history that flat grep searches miss. This Skill runs a structured, two-tier sweep that finds that residue and gates repository admission or publication on the results. ## Core Features & Use Cases - Two-tier severity scanning: BLOCKER patterns (legacy brand in filenames, content, frontmatter, manifests) gate canonical-consumer admission; ADVISORY patterns (methodology IP, personal info, operational artifacts) gate public release. - Categorized findings report: Each hit is classified as mechanical-rename, contextual-phrase, structural, binary, or manifest, with routing for remediation and a per-target summary. - Fork-event git history detection: Optional inspection of commits, tags, and branches for legacy-brand signatures, producing candidate fork-catalog entries. - Use Case: Before admitting a migrated submodule into a canonical monorepo, run the sweep in BLOCKER mode; a PASS verdict unblocks admission, while findings route mechanical renames to auto-remediation and contextual phrases to manual review. ## Quick Start Run a rebrand sweep on the target repository in report mode and show me the BLOCKER and ADVISORY findings with the Phase E admission verdict.

Frequently Asked Questions about rebrand-sweep

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find leftover brand names after rebranding a repository?

Run a structured sweep that checks folder and file names, greps content for legacy brand patterns, inspects YAML frontmatter and package manifests, and lists binary files for manual metadata review. Findings are categorized as mechanical renames or contextual phrases for remediation.

How to detect fork events from git history?

Inspect git log messages, tags, and branch names for legacy brand strings, and examine the root commit for a large initial drop suggesting a fork import. History findings are advisory by default because rewriting history is destructive.

Can the sweep automatically fix legacy brand references?

Auto-remediate mode applies sed substitutions only to findings categorized as mechanical-rename, then re-runs the sweep with before and after counts. Contextual-phrase, structural, binary, and manifest findings always require manual review.

What are the limitations of automated brand residue scanning?

Binary file content like PDF and DOCX metadata requires manual inspection, contextual-phrase classification is heuristic and can misroute findings, and git history remediation needs external tools like git filter-repo. The skill documents these as honest gaps.

When should I not run a brand residue sweep?

Skip it during mid-development on private working repos where residue is normal, and on deprecated directories that intentionally preserve historical brand context. Also avoid auto-remediate mode before reviewing a report-mode run first.