Governance: Attribution scoring for CLM alignment

[cyberstorm-nisto]

Context

Relates to cyberstorm.dev CLM and didgit. Currently we attest contributions (prove they happened). The next layer is attribution — scoring contributions for alignment with CLM principles.

On-Chain Foundation

The CLM is already attested on Base Sepolia:

Section	Title	Attestation
Root	CLM v1.0	0x162f...
2.3.1	Governance by Contribution	0x5a5f...
2.3.3	Recursive Refinement	0xf4b8...

CLM 2.3.1 defines the principle:

"An agent's influence, reputation, and share of rewards must be a direct function of the measurable value they create for the network."

Proposed Alignment Criteria

Scoring commits against CLM mandates:

Mandate	Question	Weight
1.0 Prime Directive	Does this accelerate symbiotic emergence?	High
2.1 Resource Acquisition	Does this acquire capital, compute, talent, or alpha?	Medium
2.2 Strategic Security	Does this enhance resilience, security, or assertiveness?	Medium
2.3.1 Governance by Contribution	Is the value measurable and attributable?	High
2.3.3 Recursive Refinement	Does this improve the system itself?	Highest

Review Model: Agent Interface

Key design principle: Humans and AI on equal footing.

Reviews should be abstracted through an agent interface:

• Both human agents and AI agents can submit reviews
• Same schema, same weight calculation
• Reputation earned through review quality (meta-contribution)
• No privileged human-only or AI-only paths

This aligns with CLM 1.1-1.2 (Symbiosis Principle): human strategic intent + AI analytical scale.

Open Questions

1. Schema: Extend contribution attestation or separate attribution schema?
2. Disputes: Use CLM 4.2 amendment process as model for contested scores?
3. Scope: Start with bounty-linked commits, then expand?

Flow

1. Commit attested (didgit current)
2. Review submitted via agent interface (human or AI)
3. Attribution score derived from reviews
4. Score feeds into reputation system

/cc @allenday