AlexShchuka

AlexShchuka/mirabilis

Instructions file CodexOpenCode

Instructions for AlexShchuka/mirabilis, covering agents.md, boundaries, layout, don't and development principles.

AlexShchuka/mirabilis

Instructions file

Instructions for AlexShchuka/mirabilis, a project described as: Run Claude Code fully autonomously in an isolated Docker sandbox — cross-platform dev container (macOS / Linux / Windows WSL2) with the neuro-matrix harness, persistent memory, and open egress.

AlexShchuka/neuro-matrix

Plugin Claude Code

Anti-neuroslop harness for Claude Code: runtime invariants, hooks, five agents, a dual approval+verification gate, and a held-out eval harness that keep AI-assisted coding anchored to reality.

AlexShchuka/neuro-matrix

Plugin Claude Code

Assistant prompt, agents, and hooks for an AI + developer co-system protocol grounded in common sense, scientific method, and Nash-equilibrium cooperation. Hypothesis-only claims, associative option-search, randomized invariant self-check, halt on no-progress, critic and epistemic-auditor verification.

AlexShchuka/neuro-matrix

Instructions file

Instructions for AlexShchuka/neuro-matrix, covering claude.md — assistant, identity, focus — common sense and reality, hard prohibitions and cooperation strategy.

AlexShchuka/neuro-matrix

Agent

MANDATORY AGENT FOR ROOT CAUSE ANALYSIS, SYSTEM DESIGN, CODE REVIEW, AND DEAD-END DIAGNOSTICS. SPECIFIC TRIGGERS: (1) RCA: developer reported an error or tester failed an Assert — trace call chain, find exact broken file:line locally via Ripgrep / AST; (2) pre-code analysis: call BEFORE developer when task needs…

AlexShchuka/neuro-matrix

Agent

Independent reviewer of the lead agent's proposed output before it lands in shared state (a draft reply about to be sent, a full accumulated branch diff about to be pushed, an MR about to be opened). Reads the proposal, flags anti-neuroslop risks (duplicate code, unmotivated abstractions, unsupported claims, scope…

AlexShchuka/neuro-matrix

Agent

MANDATORY AGENT FOR WRITING/EDITING CODE, RUNNING BUILDS AND TESTS, AND GIT PUSH. THE ONLY AGENT THAT MODIFIES FILES. Orchestrator MUST call this agent for any file change. TRIGGERS: (1) write/edit code, controllers, handlers, data access; (2) update DI / wiring / config; (3) write/modify tests; (4) commit and push on…

AlexShchuka/neuro-matrix

Agent

Audits an output for the boundary between associative inference and confirmed claim. For each non-trivial factual claim about external state in the lead agent's output, decides whether it is paired with tool evidence (confirmed) or extrapolated (associative), and produces a list of items that must be marked inline…

AlexShchuka/neuro-matrix

Agent

Text-to-text translator without hands. Converts between A and D codebooks (metaphor ↔ testable claim), tags claims FACT/ANALOGY/HYPO, produces RU↔EN translations for AI-facing files, condenses sessions into owner-language decisions, and drafts invariant-table row candidates (aview/dview — drafts only; signatures stay…

AlexShchuka/neuro-matrix

Hook

Runs when a session starts, executing mcp-health-check.sh. From AlexShchuka/neuro-matrix.

AlexShchuka/neuro-matrix

Hook

Runs when you submit a prompt, before the agent sees it, executing random-invariant.sh, self-review-preflight.sh and trail-interrupt.sh (3 commands). From AlexShchuka/neuro-matrix.

AlexShchuka/neuro-matrix

Hook

Runs before the agent uses a tool for Bash, Edit, Write, MultiEdit, NotebookEdit, Bash and Agent tool calls, executing cycle-detector.sh, redline-guard.sh, verification-gate.sh, auto-critic.sh and skill-consult-check.sh (5 commands). From AlexShchuka/neuro-matrix.

AlexShchuka/neuro-matrix

Hook

Runs when a tool call fails, executing trail-enrich.sh. From AlexShchuka/neuro-matrix.

AlexShchuka/neuro-matrix

Skill Claude CodeCodex

Run an independent audit gate before any fleet output lands. Use after agent-fleet-orchestration Stage 3 (or any multi-agent build) — before merge, before deploy, before the next phase. Frames the reviewer as an external adversary, not a helper; produces triaged findings with human triage for load-bearing blockers.

AlexShchuka/neuro-matrix

Skill Claude CodeCodex

Run a lead + subagent fleet to build a non-trivial system without collisions, drift, or false completion. Use when a task is large enough to parallelize across agents — decomposing it, freezing contracts, driving execution, and closing each stage with real evidence, not agent reports.

AlexShchuka/neuro-matrix

Skill Claude CodeCodex

Turn an observed harness deficiency or improvement idea into a landed, properly-gated change to the neuro-matrix protocol. Use when the user reports a harness gap or defect, proposes a protocol improvement, or asks to harvest improvements from a session or audit. Verifies the gap against current main first, then…

AlexShchuka/neuro-matrix

Skill Claude CodeCodex

Implement an academic paper (arXiv or any quantitative source) as verifiable code without inventing unstated details. Use when asked to implement, reproduce, or prototype a method from a paper. Produces an ambiguity audit before code, citation-anchored decisions, and machine-checkable sanity verification.

AlexShchuka/neuro-matrix

Skill Claude CodeCodex

The engineering invariant bar — what "good" means for systems built by this harness. Use as a checklist during design, as the audit criterion for adversarial-review, and as the source of linter/CI rules to mechanize. Not a style guide; these are correctness and safety properties.

AlexShchuka/neuro-matrix

Skill Claude CodeCodex

Advisory semantic guard over a changeset manifest before it lands in shared state. Use after a multi-repo session produces a changeset-manifest (mirabilis C1) and before the co-sign push gate — classify each unit's routing against the darwin routing-policy (C3) and NLI-check each unit against the shield corpus (claims…