Research-first context engineering workflows for Claude Code and Codex.
Use when Abu wants Codex to create, audit, or update user-level ~/.codex/AGENTS.md, repo AGENTS.md, or nested AGENTS.override.md files. Keeps instructions small, practical, and sourced from the project quality profile, routes detailed workflows to skills, and avoids stuffing research, specs, plans, or full process manuals into AGENTS.md.
Use when Abu wants Claude Code to create, audit, or update user-level ~/.claude/CLAUDE.md, repo CLAUDE.md, .claude/CLAUDE.md, CLAUDE.local.md, or .claude/rules/*.md files. Keeps instructions small, practical, and sourced from project evidence, while routing detailed workflows to skills instead of stuffing long process manuals into always-on memory.
Use when Abu wants a designer-brief.md for an AI designer or prototype agent after research, specs, stories, inspiration, or project context have been gathered. Produces a rich product and flow brief for high-fidelity mocked UI prototypes, avoids AI-slop design patterns and prescribing colors/fonts/visual style unless required, and clarifies that backend, database, wallet, auth, and API behavior should be mocked unless explicitly in scope.
Use when Abu wants a persistent Karpathy-style LLM wiki for a project or domain, including ingesting raw sources, summarizing and cross-linking concepts, updating index.md and log.md, querying accumulated knowledge, or linting stale and contradictory wiki pages.
Use when Abu wants to preserve project or session state before context gets long, before ending a work session, before switching agents, after verification, or before resuming later. Produces concise handoffs that capture objective, decisions, artifacts, current state, commands, risks, and next steps without copying full research/specs/plans into the handoff.
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.
Research-first context engineering workflows for Codex and Claude Code.
This plugin packages reusable skills for repo-local .thoughts context, reality research, project quality profiling, source-backed domain wiki maintenance, specs, stories, designer briefs, prototype discovery, research-backed plans, verification audits, handoff compaction, and instruction-file authoring.
Test directly from the local checkout:
claude --plugin-dir /Users/abu/plugins/abu-context-engineering
Install from the local marketplace:
claude plugin marketplace add /Users/abu/plugins/abu-context-engineering
claude plugin install abu-context-engineering@abu-context-engineering --scope user
/reload-plugins
Use skills with the plugin namespace:
/abu-context-engineering:reality-research
/abu-context-engineering:project-quality-profile
/abu-context-engineering:spec-writer
/abu-context-engineering:research-backed-plan
/abu-context-engineering:verification-audit
/abu-context-engineering:handoff-compaction
/abu-context-engineering:claude-md-author
If installed from GitHub:
claude plugin marketplace add Blockchain-Oracle/abu-context-engineering
claude plugin install abu-context-engineering@abu-context-engineering --scope user
The Codex manifest remains at .codex-plugin/plugin.json. The personal Codex marketplace can point at this local directory as ./plugins/abu-context-engineering.
reality-researchproject-quality-profiledomain-wikispec-writerstory-writerdesigner-briefprototype-discoveryresearch-backed-planverification-audithandoff-compactionagents-md-authorclaude-md-authorclaude plugin validate /Users/abu/plugins/abu-context-engineering
python3 /Users/abu/.codex/skills/.system/plugin-creator/scripts/validate_plugin.py /Users/abu/plugins/abu-context-engineering
npx claudepluginhub blockchain-oracle/abu-context-engineering --plugin abu-context-engineeringVerifiable agent action traces for Claude Code, backed by Sui, Walrus, Seal, and MemWal
Plugin discovery and context-aware routing system. Automatically discovers installed plugins and intelligently routes to the best tool based on task context and user preferences.
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
Harness-native ECC plugin for engineering teams - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses
Plugin-safe Claude Code distribution of Antigravity Awesome Skills with 1,642 supported skills.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.