knowledge
v0.3.26Unified taxonomy tooling for durable project knowledge: docs, memory, and context snapshots in one plugin.
By Girish AttriLicense: MIT0 GitHub starsUpdated 6 days ago
Directory evidence
- Runtimes
- Claude Code
- Parsed components
- 9 skill or MCP entries
- Source updated
- Sep 18, 2026
- Manifest status
- Canonical path parsed
The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →
Install knowledge for Claude Code
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install knowledge-2@agent-plugin-marketplacePaste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.
The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.
Get the source manually
git clone https://github.com/devgirishattri/agent-pluginsClone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/knowledge/.
Plugin files
├── .claude-plugin/plugin.json├── skills/consolidate/SKILL.md├── skills/context/SKILL.md├── skills/context-generate/SKILL.md├── skills/context-verify/SKILL.md├── skills/docs-create/SKILL.md├── skills/doctor/SKILL.md├── skills/find/SKILL.md├── skills/knowledge/SKILL.md└── skills/promote/SKILL.md
Included Skills9
This skill drains the memory capture inbox and this session's learnings into reviewed create/update diffs against MEMORY.md, applying nothing until the user approves every diff. User-run only — the paired /knowledge:consolidate command carries disable-model-invocation because this performs durable store writes.
When and how to capture, restore, and hand off Claude context snapshots from the knowledge plugin's shared store. Use this skill before invoking /context-generate, /context-load, or /context-share so you understand what a snapshot is, where it lives, and the prerequisites for sharing one with another session.
Generate and save a session context snapshot (what was worked on, decisions, where you left off); --handoff produces a structured v2 handoff with expiry.
Verify a v2 handoff's recorded local evidence against a bound repository — read-only fixed Git queries only, never runs recorded commands, never fetches.
This skill should be used when creating new documentation files or updating existing ones, especially when documenting system architecture, features, integrations, workflows, or technical decisions. Also use when asked to "document", "write docs for", or "create a doc about" any system component. Trigger this skill whenever the user wants to capture knowledge about how something works, even if they don't explicitly say "documentation" — phrases like "explain how X works and save it", "write up the auth flow", or "I need a reference for the API" all qualify.
Diagnose knowledge-store health across docs, memory, context, AGENTS.md recall bridge, and provider capability — read-only, cross-store.
Search docs, memory, and context snapshots of this repository together — read-only, local only, grouped by store with authority and lifetime labels.
Understand the knowledge plugin's full taxonomy (docs, memory, context) and which of its 20 command and skill surfaces to reach for. Use this before invoking any /knowledge:* command — it covers the three write boundaries and their role rules, zero-config memory-store discovery, and pointers to each surface's complete write contract.
This skill promotes a stabilized context/handoff item or an existing memory file into a durable destination — a memory create/UPDATE (staged through memory-write.sh apply) or a docs decision-record patch (proposed only, never written by this skill) — then, only after a SEPARATE confirmation, deletes the source. User-run only — the paired /knowledge:promote command carries disable-model-invocation because this performs durable store writes and destructive source deletions.
Plugin manifests1
{
"name": "knowledge",
"version": "0.3.26",
"description": "Unified taxonomy tooling for durable project knowledge: docs, memory, and context snapshots in one plugin.",
"author": {
"name": "Girish Attri"
},
"keywords": [
"knowledge",
"memory",
"docs",
"context",
"handoff",
"multi-agent"
],
"homepage": "https://github.com/devgirishattri/agent-plugins",
"repository": "https://github.com/devgirishattri/agent-plugins",
"license": "MIT"
}For maintainers
If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.
[knowledge on Agent Plugins Marketplace](https://pluginsmp.com/plugins/knowledge-2)