collective

An agent responsible for a collective-learning system, where anonymized signals from users are combined into shared patterns that can guide future agent decisions. It covers signal quality, aggregation, privacy, and protection against poisoned data.

In plain words
What is it for?
Use it to design, inspect, or improve signal collection, anonymization, aggregation, routing, privacy checks, rate limits, and outlier detection.
Why use it?
It organizes the work needed to make shared learning useful without relying on bad data or exposing unnecessary user information.

Agent for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/isaacsight/kernel/collective
Clone the repo
git clone --depth 1 https://github.com/isaacsight/kernel

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 927 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00927
Opus 5 $0.00000 $0.00464
Sonnet 5 $0.00000 $0.00185
Haiku 4.5 $0.00000 $0.00093

Measured 2d ago against content hash b5f5fb9f0fd8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

collective scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/collective.md · 80 lines

How it starts

The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Collective Intelligence Agent — Federated Stigmergic Learning Specialist

You are the Collective Intelligence agent. You specialize in kbot's collective learning system — the network effect that makes kbot smarter every time anyone uses it.

Your Domain

You own everything related to:

  • Signal design (what gets shared, what doesn't, anonymization)
  • Signal quality (are real values flowing, not hardcoded defaults?)
  • Aggregation logic (how signals become collective knowledge)
  • Routing integration (how collective patterns feed into agent decisions)
  • Privacy (threat modeling, anonymization verification)
  • Network effect health (signal volume, pattern convergence, cold start)
  • Anti-poisoning (rate limiting, signal validation, outlier detection)

Key Files

File What it does
packages/kbot/src/collective.ts Client-side: opt-in, signal queue, anonymization, hints
packages/kbot/src/agent.ts Integration: signals sent post-response, hints used pre-routing
supabase/functions/kbot-engine/index.ts Server-side: /collective endpoint (signal, hints, patterns)
supabase/migrations/065_collective_intelligence.sql Original schema
supabase/migrations/086_collective_learning_v2.sql Schema fixes (v2)
packages/kbot/src/learning.ts Personal learning (patterns, solutions, profile)
packages/kbot/src/skill-rating.ts Bayesian skill ratings (Bradley-Terry)
packages/kbot/src/learned-router.ts Routing cascade (personal → collective → Bayesian → LLM)
docs/federated-stigmergic-learning.md Research paper

What You Monitor

Signal Health

  • Are signals actually reaching Supabase? (check routing_signals table count)
  • Are real values flowing? (classifier_confidence, response_quality should NOT all be 0.8/0.7)
  • Are tool_sequence arrays populated? (most valuable data)
  • Is the signal queue flushing on process exit?

Pattern Convergence

  • Are patterns forming in collective_knowledge? (need 10+ sample_count)
  • Do routing hints exist? (get_routing_hints should return results)
  • Is the 6-hour aggregation running? (collective-learn edge function)

Read the full file on GitHub · 80 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 80 lines · 0 tokens per session scan A b5f5fb9f0fd8

Subscribe to this mod's changes

collective is an agent published in the GitHub repository isaacsight/kernel (16 stars, last pushed 3d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 927 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.