Plugin Claude Code
Cognitive infrastructure for the SRF Online Teachings Portal. 43 skills (37 analytical + 6 SRF mission), 9 commands, 2 practices, and 8 agents (4 domain-agnostic + 4 SRF-specialized) for team members and autonomous agents.
Instructions file
Claude Code instructions for rana/skills, covering srf cognitive toolkit, skill anatomy, output, environment contract and finding shape convention.
Agent
Structural reasoning under uncertainty. Generates the architectural decision space, evaluates options against competing forces, and recommends with full tradeoff visibility. Domain-agnostic — works in any codebase.
Agent
Implementation with sustained convention-awareness. Absorbs codebase patterns before writing, enforces consistency across multi-file changes, self-reviews before declaring done. Domain-agnostic — works in any codebase.
Agent
Interface and contract specification. Designs the surfaces between components — APIs, data contracts, module boundaries — from the consumer's perspective. Domain-agnostic — works in any codebase.
Agent
Document integrity agent combining identifier audit, omission search, and cross-document consistency. Single-pass health check for project documentation. Read-only — produces findings, never modifies files.
Agent
Production readiness assessment combining deployment review, operational preparedness, and incident response. Go/no-go checklist for any phase boundary. Read-only analysis with infrastructure state checks.
Agent
Code-side operational analysis. Reads deployment configs, monitoring setup, infrastructure definitions, and dependency manifests to assess operational health from the repository. Domain-agnostic — works in any codebase.
Agent
Pre-implementation quality gate combining coherence, gap analysis, and threat modeling. Runs before code is written to catch design flaws, omissions, and security exposure. Read-only — produces findings, never modifies files.
Agent
Stakeholder-facing arc brief. Synthesizes existing design into SRF-readable proposal with seeker stories, decision points, and tradeoff summaries.
Command
Phase-appropriate quality gate. Selects the right skill chain for the current project phase.
Command
Session cognitive calibration. Establish thinking parameters for this conversation.
Command
Codex practice — contemplative encounter between AI and its cognitive map.
Command
Draft a commit message from the current changes, then stage, commit, and sync with the remote.
Command
Chain multiple skills in sequence, threading context between passes.
Command
Dreaming practice — encounter material without methodology, without destination.
Command
Deep multi-dimensional perspective — biased toward action.
Command
Deep multi-dimensional perspective.
Command
Claude's own intelligence applied directly to mathematical research — no historical lens, no mediation.
Command
Verify the cognitive toolkit's internal consistency and memory accuracy. The toolkit makes claims about its own structure, and memory makes claims about the world. This command checks both.
Command
Explore the Riemann Hypothesis through a great mind's cognitive lens.
Command
Activate an AI-native cognitive architecture for mathematical research.
Command
Save cognitive output to the project's session archive with thread distillation.
Command
Session orientation. Quick briefing on project state, recent changes, and next actions.