By pwarnock
Progressive disclosure methodology for LLM-consumed documents — organize any doc (CLAUDE.md, AGENTS.md, system prompts, runbooks) using three-tier lazy loading
Organize any LLM-consumed document using three-tier progressive disclosure (lazy prompting). Keeps high-signal content always loaded, defers low-frequency detail to on-demand files.
Add to your Claude Code plugins:
claude plugin add /path/to/progressive-disclosure
Or install from the marketplace.
/progressive-disclosure
Runs an interactive workflow that:
When you edit a target document past a line threshold (200 for project-level, 100 for user-level), the plugin suggests running /progressive-disclosure.
Edit config/targets.txt to add custom glob patterns:
*/CLAUDE.md
*/.claude/CLAUDE.md
*/AGENTS.md
*/system-prompt.md
docs/runbook-*.md
| Tier | When Loaded | Examples |
|---|---|---|
| Level 1 | Every session | Core commands, hard rules, code patterns, error diagnostics |
| Level 1.5 | When conditions match | .claude/rules/ (CLAUDE.md), conditional sub-docs (AGENTS.md) |
| Level 2 | On demand via trigger | SOPs, edge cases, historical decisions, full config examples |
/revise-claude-md (from claude-md-management plugin) — routes new learnings to the correct tier, preventing re-bloatprogressive-disclosure, lazy-prompting, progressive-discovery, claude-md, agents-md, optimization
Modifies files
Hook triggers on file write and edit operations
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub pwarnock/pwarnock-cc-plugins --plugin progressive-disclosureThree-tier persistent context infrastructure for AI coding agents — constitution, routing intelligence, and cold memory. Based on 'Codified Context: Infrastructure for AI Agents in a Complex Codebase' (Vasilopoulos, 2025).
Session retrospective skill with hook-driven prompting. Generates structured session summaries with learnings that feed back into project memory.
Capture meetups, contacts, and interactions from natural language into Notion CRM
Prisma 7 migration and best practices skills for Claude Code
Bridge Gastown cross-session orchestration with in-session subagent parallelization
Optimize CLAUDE.md and docs/ ecosystem following Boris Cherny's best practices
Optimize your CLAUDE.md hierarchy — deduplicate, detect misplacement, compress bloat, and visualize token usage across global/project/repo levels
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
CLAUDE.md lifecycle toolkit: initialise, enhance, sync, and modularise CLAUDE.md files with a 150-line cap, automatic chaining, and Karpathy behavioural guardrails.
Rewrite a CLAUDE.md file using <important if> blocks to improve instruction adherence
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.