By NeoHiveAI
NeoHive cognitive memory: MCP server registration, managed rules for tool usage, the explore-neohive subagent, a PreToolUse nudge on Glob/Grep, end-user-clear skill names (load-context, capture-session-learnings, design-codebase-docs, enable-smart-prompts, generate-claude-md), and post-session learning extraction
Use when the user says "save what I learned", "capture this session", "extract learnings", or at the end of any session (also fires automatically from the stop hook). Scans the conversation transcript, extracts corrections, conventions, decisions, and gotchas worth persisting, deduplicates against NeoHive, stores up to 5 new memories, and audits CLAUDE.md for instruction gaps.
Use when the user says "generate docs", "build out documentation", "document this codebase", "design our doc standard", or "NeoHive should help with docs". Multi-phase orchestrator that designs a documentation standard tailored to the codebase through Socratic dialogue (audience, depth, reference vs tutorial, per-boundary vs per-subservice), saves it to NeoHive as a reusable convention, generates 2-3 sample pages to validate the approach, then hands off to a fresh session for full generation. Output is the agreed standard plus validated samples, not a complete doc set.
Use when the user says "make NeoHive smarter", "rewrite my prompts before searching memory", "enable smart recall", "give me a smarter hook", or during `/neohive:getting-started` Phase 4. Generates a tailored UserPromptSubmit hook that uses a small model (Haiku by default) to rewrite the user's prompt into a good `memory_recall` query, decides when lookup is worthwhile, calls NeoHive, and injects only the most relevant results. Replaces the default plugin hook which passes prompts verbatim.
Generate a project-specific NeoHive topology block in ./CLAUDE.md by surveying connected hives (list_hives + memory_stats + sampled memory_recall probes). Runs as Phase 3 of /neohive:getting-started, and is also user-invocable for re-runs when hives change. Use when the user says "generate my NeoHive CLAUDE.md", "regenerate the topology", "re-survey my hives", or after adding/removing/renaming hives.
DEPRECATED alias for `/neohive:design-codebase-docs`. This slug will be removed in a future minor release. Use `/neohive:design-codebase-docs` instead.
Uses power tools
Uses Bash, Write, or Edit tools
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.
A Claude Code plugin marketplace for the NeoHive cognitive memory system. Install the neohive plugin to wire Claude Code into any NeoHive MCP server — persistent semantic memory across sessions, automatic context recall, and post-session learning extraction.
/plugin marketplace add NeoHiveAi/NeoHiveClaude
/plugin install neohive@neohive-claude
/neohive:getting-started
One command walks you through verifying the MCP server, setting up auth, generating a project-specific topology block in your CLAUDE.md, migrating any existing project memory (CLAUDE.md, AGENTS.md, .claude/rules) into NeoHive, and optionally enabling the smart-recall hook. 3–5 minutes end-to-end.
If your NeoHive server requires auth, export a bearer token before launching Claude:
export NEOHIVE_TOKEN="your-token-here"
Disable the auto-context hook for a session:
export NEOHIVE_HOOK_DISABLED=1
neohiveNeoHive cognitive memory — MCP server registration, managed rules for tool usage, and post-session learning extraction.
| Type | Name | Description |
|---|---|---|
| Skill | getting-started | Guided first-run: verify MCP, set auth, generate topology block, migrate memory, enable helpers. Start here. |
| Skill | load-context | Pre-load relevant NeoHive memories for the current task via memory_context. Run at the start of every session. |
| Skill | generate-claude-md | Survey connected hives and write a project-specific topology block into ./CLAUDE.md (hive table, write-routing, session-start non-negotiables). Re-runnable when hives change. |
| Skill | migrate-memory | Scan local CLAUDE.md / AGENTS.md / .claude/rules and migrate project-scoped entries into NeoHive |
| Skill | design-codebase-docs | Design a documentation gold standard through Socratic dialogue, save to NeoHive, validate with sample pages |
| Skill | enable-smart-prompts | Generate a tailored smart-recall hook that rewrites prompts with a small model before querying NeoHive |
| Skill | capture-session-learnings | End-of-session extraction of learnings, corrections, and insights into NeoHive semantic memory |
| Agent | explore-neohive | Semantic-first codebase + knowledge exploration subagent. Prefer this over the built-in Explore in NeoHive-indexed projects. |
| Hook | SessionStart | Installs/updates ~/.claude/rules/neohive.md with persistent tool-usage instructions |
| Hook | UserPromptSubmit | Injects relevant memories into context automatically on every prompt |
| Hook | PreToolUse (Glob, Grep) | Nudges Claude toward memory_recall when running broad filesystem searches inside an indexed project (NEOHIVE_PRETOOL_STRICT=1 to hard-deny, NEOHIVE_PRETOOL_DISABLED=1 to opt out) |
The slugs start, revise-vector-memory, generate-docs, and generate-post-submit-hook remain as deprecated aliases that redirect to the new names; they will be removed in a future minor release.
The plugin does not ship a pre-configured MCP server — you register your own NeoHive gateway via Claude Code's built-in MCP commands (e.g. /mcp in-session or claude mcp add). The hook and skills discover any MCP server whose key contains neohive in your project .mcp.json or ~/.claude.json, so one registration covers the whole plugin.
NeoHiveClaude/
├── .claude-plugin/
│ └── marketplace.json # Marketplace catalog
└── plugins/
└── neohive/
├── .claude-plugin/plugin.json
├── agents/
│ └── explore-neohive.md # Semantic-first exploration subagent
├── hooks/
│ ├── hooks.json
│ ├── session-start.sh # Manages ~/.claude/rules/neohive.md
│ ├── neohive-context.sh # UserPromptSubmit memory recall
│ └── pretool-tree-walker.sh # PreToolUse nudge on Glob/Grep
├── rules/neohive.md # Persistent tool-usage instructions
└── skills/
├── getting-started/SKILL.md
├── load-context/SKILL.md
├── capture-session-learnings/SKILL.md
├── generate-claude-md/SKILL.md
├── migrate-memory/SKILL.md
├── design-codebase-docs/SKILL.md
├── enable-smart-prompts/
│ ├── SKILL.md
│ └── template.sh
├── start/SKILL.md # deprecated alias
├── revise-vector-memory/SKILL.md # deprecated alias
├── generate-docs/SKILL.md # deprecated alias
└── generate-post-submit-hook/SKILL.md # deprecated alias
Test locally without installing:
claude --plugin-dir ./plugins/neohive
npx claudepluginhub neohiveai/neohiveclaude --plugin neohiveClaude + Obsidian knowledge companion. Sets up a persistent, compounding wiki vault (Karpathy's LLM Wiki pattern). v1.7 "Compound Vault" + v1.8 methodology modes close 5 of 5 priority gaps from the May 2026 compass artifact. Ships: substrate alignment with kepano/obsidian-skills, default Obsidian CLI transport, hybrid retrieval (contextual prefix + BM25 + cosine rerank per Anthropic's Sept 2024 research), per-file advisory locking for multi-writer safety, pre-commit verifier agent, AND methodology modes (LYT / PARA / Zettelkasten / Generic) for first-class organizational support no other Claude+Obsidian competitor offers. v1.7.x audit closure: every BLOCKER + HIGH + MEDIUM + LOW finding from the v1.7.0 audit is CLOSED or DEFERRED-with-rationale. Optional DragonScale Memory extension (log folds, deterministic addresses, semantic tiling lint, boundary-first autoresearch).
Complete AI coding workflow system. Self-correcting memory + persistent FTS5-indexed research wikis + auto-research loop + multi-LLM council on a single SQLite store. 33 skills, 8 agents, 22 commands, 37 hook scripts across 24 events. Cross-agent via SkillKit.
Make your AI agent code with your project's architecture, rules, and decisions.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
AI-powered wiki generator for code repositories. Generates comprehensive, Mermaid-rich documentation with dark-mode VitePress sites, onboarding guides, deep research, and source citations. Inspired by OpenDeepWiki and deepwiki-open.