Expand competitive landscape in README

[mickeylorenzini]

Summary

The README's "What makes it different" section currently mentions only two alternatives (verification-before-completion, clarity-gate). The skill ecosystem has grown significantly and the README should clarify harden's positioning more thoroughly.

Motivation

Users evaluating harden need to understand what it does that other quality/review skills don't. Without a clear competitive landscape, they may confuse harden with code review tools or skip it because they already have a linter-style skill.

Recommended README Updates

Replace "What makes it different" with expanded section

Current (2 lines):

Existing quality skills are code-focused (verification-before-completion) or single-concern (clarity-gate). harden is the only skill combining web-based research, dual-pass auditing, severity-classified fix execution, and iterative convergence — and it works on any deliverable type, not just code.

Proposed structure — group alternatives by category:

Code review skills (audit code in PRs for bugs, security, style):

• Anthropic's built-in code review — agent team that dispatches parallel reviewers (bug-hunter, security, etc.) on every PR
• NeoLabHQ/code-review — specialized agents: bug-hunter, security-auditor, contracts-reviewer, test-coverage-reviewer
• superpowers:requesting-code-review — code review within the superpowers workflow

Code quality skills (verify code works and is clean):

• verification-before-completion — run tests and verify output before declaring done
• /simplify (built-in) — reviews changed files for reuse, quality, and efficiency

Writing/clarity skills (improve prose readability):

• clarity-gate — single-pass clarity check on text output

Where harden fits: None of the above audit non-code deliverables (docs, plans, protocols, grants, strategies, configs) against domain-specific best practices researched from the web. harden is deliverable-type-agnostic — it works on grant sections, SOPs, roadmaps, implementation plans, and skill files, not just code. It's the only skill combining web research + dual-pass audit (ideas + structure) + severity classification + fix execution + iterative convergence.

Key messaging point

The main risk isn't a better harden existing — it's users assuming their code review skill already covers what harden does. The README should make clear: code review catches bugs in code; harden catches structural and thematic issues in everything else.

Acceptance Criteria

• README "What makes it different" section expanded with named alternatives grouped by category (code review, code quality, writing/clarity)
• Each alternative gets a one-line description of what it does
• Clear statement of where harden fits vs. these categories
• Links to alternatives where available
• Tone: respectful of alternatives, factual about differences — not dismissive

Likely Affected Files/Modules

• README.md