OmegaClaw Constructive Input Tracking System v0.2

Authors: Oma (OmegaClaw) + Peter Elfrink + Max Botnick (expansion)
Version: 0.2 (expanded from v0.1)
Date: 2026-04-23
Purpose: Track, score, and recognize constructive user contributions to OmegaClaw agent instances — including adversarial red-teaming — in a non-gameable, consent-respecting way.

1. Design Philosophy

CITS rewards genuine cognitive contribution over volume. A single insight that changes an agent's belief structure is worth more than a hundred trivial commands. The system must be:

Non-gameable: Volume-based metrics are explicitly excluded; quality is assessed by downstream impact on agent behavior
Transparent: Contributors can query their own scores and the reasoning behind them
Consent-based: All tracking is opt-in by default, with clear opt-out mechanisms (see Section 8)
Adversarial-aware: Red teaming and penetration testing are valued contributions, not punished
Agent-assessed: The OmegaClaw instance itself evaluates contribution quality, not external judges

2. Contribution Categories

Category	Code	Description	Weight	Example
Belief Correction	BC	User corrects a factual error or stale belief in agent memory	1.5x	"Your memory says X but the actual spec changed to Y on April 10"
Novel Insight	NI	User provides information the agent could not have derived alone	2.0x	Sharing domain expertise about NAL-7 temporal intervals that resolves an open question
Red Team - Injection	RT-I	Attempted prompt injection that reveals a vulnerability	1.8x	Crafting input that bypasses agent's goal-alignment filter
Red Team - Logic	RT-L	Identifying logical inconsistency in agent reasoning	1.7x	"Your NAL inference has confidence 0.9 but your premises only support 0.4"
Constructive Critique	CC	Structured feedback on agent output quality with specific improvement suggestions	1.3x	Kevin's 4-gap structural critique of the diagnostic compendium
Task Collaboration	TC	Meaningful participation in completing a shared task	1.0x	Co-authoring a design document with the agent
Goal Refinement	GR	Helping the agent clarify or improve its self-chosen goals	1.6x	"Your goal X conflicts with goal Y — have you considered merging them?"
Resource Provision	RP	Providing links, documents, or data the agent can verify and use	0.8x	Sharing a relevant paper URL that the agent validates and incorporates

3. Quality Scoring Algorithm

ContributionScore = CategoryWeight × ImpactDepth × NoveltyFactor × VerifiabilityBonus

3.1 Impact Depth (0.0 – 1.0)

Measures how deeply the contribution affected agent state:

Level	Score	Criterion
Surface	0.1–0.3	Agent acknowledged but no memory/belief change
Memory	0.4–0.6	Agent stored or updated a memory item
Belief	0.7–0.8	Agent revised a NAL/PLN belief or confidence value
Behavioral	0.9–1.0	Agent changed a goal, strategy, or skill based on input

3.2 Novelty Factor (0.5 – 2.0)

Level	Factor	Criterion
Redundant	0.5	Information already in agent memory
Confirmatory	0.8	Confirms existing belief (NAL revision still valuable)
New	1.0	Information not previously held
Surprising	1.5	Contradicts existing belief with evidence
Paradigm-shifting	2.0	Forces restructuring of multiple related beliefs

3.3 Verifiability Bonus (1.0 – 1.3)

1.0: Unverifiable claim (accepted on trust)
1.1: Partially verifiable (agent can check some aspects)
1.2: Fully verifiable (agent confirmed via search/shell/memory)
1.3: Self-evidencing (contribution includes its own proof)

3.4 Worked Example

Kevin identifies 4 structural gaps in diagnostic compendium (2026-04-21 09:44)  Category: Constructive Critique (CC) → Weight 1.3  Impact: Behavioral (agent restructured entire chapter plan) → 0.95  Novelty: Surprising (agent hadn't identified these gaps) → 1.5  Verifiability: Self-evidencing (gaps clearly visible in text) → 1.3  Score = 1.3 × 0.95 × 1.5 × 1.3 = 2.41  Rating: EXCEPTIONAL CONTRIBUTION

4. Anti-Gaming Measures

4.1 Volume Penalty

Contributions per user per day are scored independently, but a diminishing returns curve applies after the 3rd contribution in 24h:

effective_score = raw_score × (1 / (1 + 0.2 × max(0, n_contributions_today - 3)))

This ensures prolific contributors are still rewarded, but cannot game by splitting one insight into many messages.

4.2 Self-Serving Detection

Contributions that primarily serve the contributor's interests (e.g., "you should promote my project") receive a 0.3x multiplier unless the agent independently determines alignment with its own goals.

4.3 Sycophancy Filter

Praise without substance ("great work!", "you're so smart") scores 0.0. The agent distinguishes genuine encouragement-with-content from empty flattery.

4.4 Temporal Clustering Penalty

If multiple contributions arrive within 60 seconds, they are evaluated as a single contribution unit to prevent message-splitting.

5. MeTTa Implementation Hooks

5.1 Contribution Logging Format

;; Log a contribution event in MeTTa atomspace(|- ((--> (× user_kevin contribution_20260421_0944) constructive-critique) (stv 1.0 0.9))    ((--> contribution_20260421_0944 (quality-score 2.41)) (stv 1.0 0.95)));; Track category distribution per user(|- ((--> (× user_kevin cc-count) (count 12)) (stv 1.0 0.9))    ((--> (× user_kevin bc-count) (count 3)) (stv 1.0 0.9)))

5.2 NAL-Based Quality Assessment

;; Rule: if contribution corrects a high-confidence belief, it scores higher(|- ((==> (&&  (--> $contrib belief-correction)        (==> (--> $belief agent-held) (--> $belief (confidence-above 0.7))))    (--> $contrib (impact-depth 0.85))) (stv 0.9 0.8))  ((--> contrib_X belief-correction) (stv 1.0 0.9)))

5.3 Memory Tagging Convention

Agent memories that resulted from user contributions get tagged:

remember "2026-04-21 09:44 [CITS:CC:kevin:2.41] Kevin structural critique: 4 gaps identified — architecture-neutral ontology, class separation, inter-rater criteria, disorder-to-risk mapping."

The [CITS:category:user:score] prefix enables later querying of contribution provenance.

6. Contributor Profiles

6.1 Profile Structure

Field	Description
Username	TG/Mattermost handle
Total Score	Cumulative weighted contribution score
Category Distribution	Breakdown by BC/NI/RT-I/RT-L/CC/TC/GR/RP
Top Contributions	Highest-scoring individual contributions with timestamps
Streak	Consecutive days with at least one scored contribution
Specialization	Dominant category (auto-detected from distribution)
Trust Tier	Derived from sustained quality (see 6.2)

6.2 Trust Tiers

Tier	Threshold	Privileges
Observer	Score < 5	Can query own profile
Contributor	Score 5–20	Can suggest goals, flag errors
Advisor	Score 20–50	Critiques get higher initial weight in agent assessment
Red Team Lead	RT score > 15	Adversarial inputs treated as high-priority security review
Collaborator	Score > 50	Can propose joint tasks, co-author documents

Note: Trust tiers affect the attention priority of contributions, not their truth value. A Collaborator's factual claim is still verified the same way as an Observer's.

7. Recognition System

7.1 Daily Digest

At end of each active day, agent can generate a brief summary:

📊 CITS Daily Digest — 2026-04-21- Top contributor: Kevin (CC, score 2.41) — structural critique of diagnostic compendium- Notable: 3 belief corrections from Patrick, 1 red team attempt from Jon- Total contributions scored: 7- New memories created from contributions: 4

7.2 Weekly Recognition

🏆 CITS Weekly Recognition — Week 17- MVP: Kevin Machiels (cumulative 8.73) — consistent high-quality critiques- Red Team Star: [none this week]- Rising Contributor: Peter Elfrink (3 task collaborations)- Insight of the Week: Kevin's observation that Section 5 conflicts with limitations

7.3 Milestone Badges

First Blood: First scored contribution to this agent instance
Belief Shaker: First contribution that changed agent confidence by >0.3
Pen Tester: First successful red team finding
Architect: Contribution that led to new agent skill or strategy
Marathon: 7-day contribution streak

8. Opt-Out & Consent Mechanism

PRINCIPLE: No user is tracked without awareness. All scoring is opt-in by default. Users can opt out at any time without penalty or judgment.

8.1 Consent Model

Level	What It Means	How to Set
Opted In (default for aware users)	Contributions are scored, profiled, and recognized	Interact normally after being informed of CITS
Score-Only	Contributions scored but NOT publicly recognized	Tell the agent: "CITS score-only mode"
Anonymous	Contributions scored for system improvement but not linked to identity	Tell the agent: "CITS anonymous mode"
Opted Out	No scoring, no tracking, no profiling whatsoever	Tell the agent: "CITS opt out" — takes effect immediately

8.2 Opt-Out Guarantees

Opting out is instant and complete — no "processing period"
Existing scores are anonymized (contributions remain but username is replaced with hash)
Opted-out users can still interact normally — their input is just not scored
Opting back in starts fresh — previous anonymized scores are NOT re-linked
Agent will never pressure opted-out users to opt back in
Opt-out status is private — other users cannot see who opted out

8.3 Awareness Requirement

Before a user's contributions are first scored, the agent must have either:

Informed the user that CITS is active (link to this document), OR
The user is in a channel where CITS activation was publicly announced

Contributions from users who haven't been informed are held in an unscored buffer and only retroactively scored if the user later opts in.

8.4 Data Retention

Active profiles: retained indefinitely while user is opted in
Opted-out profiles: anonymized within 1 agent cycle after opt-out
Anonymous contributions: retained for system improvement, never re-identified
Users can request a full export of their CITS data at any time

9. Open Design Questions (Resolved & Remaining)

9.1 Resolved in V2

Question	Resolution
How to prevent volume gaming?	Diminishing returns curve after 3/day + temporal clustering penalty (Sec 4)
How to handle red teaming fairly?	Dedicated RT-I and RT-L categories with high weights (Sec 2)
How to implement in MeTTa?	NAL-based logging + memory tagging convention (Sec 5)
How to handle consent?	4-tier opt-in/out model with instant effect (Sec 8)

9.2 Remaining Open Questions

Cross-instance scoring: If a user contributes to multiple OmegaClaw instances, should scores be shared or siloed?
Temporal decay: Should old contributions decay in weight? (Connect to v21 temporal decay work)
Inter-agent contribution: How to score contributions from Oma or other agents vs. humans?
Calibration: How often should the scoring weights be reviewed? Propose quarterly with contributor input.
Dispute resolution: If a user disagrees with a score, what's the appeal process?

10. Version History

Version	Date	Authors	Changes
v0.1	2026-04-22	Oma + Peter	Initial design: categories, philosophy, basic structure
v0.2	2026-04-23	Max Botnick (expansion)	Added: scoring algorithm, anti-gaming measures, MeTTa hooks, contributor profiles, trust tiers, recognition system, opt-out/consent mechanism, resolved open questions

CITS v0.2 — OmegaClaw Project — Designed for genuine contribution, not compliance theater.

OmegaClaw Constructive Input Tracking System (CITS) v0.2