feat: Fix critical blockers and frontmatter compliance for skill-bench artifacts

Part of the skill-bench integration epic: #1504

Before the bench framework can land in hve-core, 3 critical and 12 major conformance issues identified in the 27-finding quality review must be resolved.

Critical (3):

• Add the missing subagent file referenced by the orchestrator
• Fix workspace naming mismatch (path alignment with .copilot-tracking/)
• Add missing agents: frontmatter to all agent files

Major (12):

• Split SKILL.md to bring it under the 500-line limit
• Add missing name: frontmatter to agents
• Replace Unix-only shell commands with cross-platform equivalents (PowerShell + bash)
• Normalize non-standard section headers to match hve-core conventions

Acceptance criteria:

• All agent files pass validate:skills without errors
• No Unix-only commands remain in scripts
• agents: frontmatter present on all .agent.md files
• SKILL.md is ≤ 500 lines
  Related to #1504

[!NOTE]

🔒 Integrity filter blocked 2 items

The following items were blocked because they don't meet the GitHub integrity level.

• microsoft/hve-core#1498 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• microsoft/hve-core#1482 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

Generated by Issue Triage for issue #1504 · ● 859.8K · ◷

Start with the skill-bench integration epic in #1504 and inspect the referenced agent files, SKILL.md, scripts, and .copilot-tracking paths. Run validate:skills first, then check the agent frontmatter, workspace paths, section headers, and shell commands. Done means validation passes, SKILL.md is at most 500 lines, all .agent.md files have the required frontmatter, and no Unix-only commands remain.