Assess Signalproof harness and skill streamlining

[docreo]

Purpose

Add an evidence-backed September assessment for reducing redundant Skill and command overhead while preserving Signalproof authority, verification, recovery, and continuity.

Key findings

• 39 Active Skills, 28 Active commands, 10 registered Loops, and 3 Candidate capability families now require normalized runtime-use measurement.
• Registry presence and acceptance evidence do not prove runtime invocation.
• Nested coordinator Skills and compatibility commands are the clearest consolidation targets.
• Rare security and recovery controls should remain lazy and narrow rather than being retired by frequency alone.
• Rico and future builders should carry a lightweight operating mentality and retrieve task-specific knowledge through the same harness, Git, tools, tests, and evidence loop.
• The CLI/Harness should record privacy-minimized Skill selection receipts and run with/without-Skill acceptance comparisons before retirement.

Scope

Documentation only. No Active Skill, command, registry, Router, Loop, or lifecycle status changes.

Verification

• 102 consistency tests: PASS
• 5 public-boundary security tests: PASS
• suite consistency: PASS
• Library consistency: PASS
• Skill budget: PASS
• public-boundary check: PASS
• Git blob identity verified after push

Next gate

Owner review of the assessment before any vNext simplification candidate changes Active capability.