trapic-mcp 0.1.0

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Trapic
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,152 @@
+# Trapic MCP
+
+> AI's long-term memory layer — capture, search, and connect knowledge across AI coding sessions.
+
+Trapic is an MCP (Model Context Protocol) server that automatically records decisions, facts, conventions, and discoveries from your AI conversations. Knowledge persists across sessions and devices.
+
+## Why Trapic?
+
+Every day you make dozens of decisions with AI — why this framework, how that bug was fixed, what conventions the team follows. **All of it disappears when the session ends.**
+
+Trapic captures this knowledge automatically and makes it searchable, connected, and persistent.
+
+## Quick Start
+
+### 1. Get your API token
+
+Sign up at [trapic.ai](https://trapic.ai) and create an API token.
+
+### 2. Install
+
+```bash
+npm install -g trapic-mcp
+```
+
+### 3. Login
+
+```bash
+trapic login
+```
+
+### 4. Connect to your AI tool
+
+**Claude Code** — Add to `~/.claude/mcp.json`:
+
+```json
+{
+  "mcpServers": {
+    "trapic": {
+      "type": "url",
+      "url": "https://mcp.trapic.ai/mcp",
+      "headers": {
+        "Authorization": "Bearer YOUR_TOKEN"
+      }
+    }
+  }
+}
+```
+
+**Cursor** — Add to `.cursor/mcp.json`:
+
+```json
+{
+  "mcpServers": {
+    "trapic": {
+      "url": "https://mcp.trapic.ai/mcp",
+      "headers": {
+        "Authorization": "Bearer YOUR_TOKEN"
+      }
+    }
+  }
+}
+```
+
+**Other platforms** — See [full setup guide](https://trapic.ai/docs).
+
+### 5. Add instructions to your project
+
+Create a `CLAUDE.md` (or `.cursorrules`, etc.) in your project root:
+
+```markdown
+## Auto-load Knowledge
+At the start of every conversation, read the MCP resource
+`trapic://knowledge/project/your-project` to load existing knowledge.
+
+## Knowledge Capture
+When working on this project, automatically record important knowledge
+using `trapic_create`:
+- **decision**: Technical choices (why X over Y)
+- **fact**: Non-obvious discoveries
+- **convention**: Established patterns
+- **state**: Project milestones
+
+Call trapic_create with: content, context, type, tags: ["project:your-project", ...]
+```
+
+## MCP Tools
+
+| Tool | Description |
+|------|-------------|
+| `trapic_create` | Record a knowledge entry with auto-contextualize and conflict detection |
+| `trapic_search` | Semantic search across your knowledge base |
+| `trapic_update` | Update a trace's content, status, or metadata |
+| `trapic_get_chain` | Retrieve the causal chain upstream from a trace |
+| `trapic_get_effects` | Find traces caused by a given trace |
+| `trapic_get_active` | Get all active traces, optionally filtered by tags |
+| `trapic_get_context` | Get a context cluster with its related traces |
+| `trapic_find_contexts` | Search for relevant context clusters |
+| `trapic_contextualize` | Assign a trace to context clusters using LLM |
+| `trapic_extract` | Extract structured knowledge from text |
+| `trapic_auto_summary` | Batch extract traces from a conversation summary |
+
+## MCP Resources
+
+| Resource | Description |
+|----------|-------------|
+| `trapic://knowledge/active` | All active knowledge (conventions, preferences, states) |
+| `trapic://knowledge/project/{name}` | Project-specific knowledge |
+
+## Features
+
+- **Zero-friction capture** — AI records knowledge during work, no manual entry
+- **Semantic search** — Find knowledge by meaning, not keywords
+- **Auto-contextualize** — New knowledge automatically linked to related clusters
+- **Conflict detection** — Contradictions detected and flagged or auto-resolved
+- **Causal chains** — Trace how decisions led to other decisions
+- **Cross-platform** — Works with Claude Code, Cursor, Windsurf, Cline, Copilot, Gemini
+- **Field-level encryption** — Content encrypted at rest with per-user keys
+- **Team collaboration** — Share knowledge across team members
+
+## CLI Commands
+
+```bash
+trapic login          # Authenticate with your Trapic account
+trapic logout         # Clear saved credentials
+trapic whoami         # Show current user info
+trapic status         # Show connection status
+trapic search <query> # Search your knowledge base
+trapic list           # List recent traces
+trapic init           # Initialize project with rules files
+trapic token          # Manage API tokens
+```
+
+## Self-hosting
+
+Trapic MCP can be self-hosted. You'll need:
+
+- Supabase project (PostgreSQL + Auth)
+- OpenAI API key (for embeddings)
+- Groq API key (optional, for LLM features)
+
+See the [self-hosting guide](https://trapic.ai/docs) for details.
+
+## License
+
+MIT — see [LICENSE](./LICENSE)
+
+## Links
+
+- [Website](https://trapic.ai)
+- [Documentation](https://trapic.ai/docs)
+- [GitHub](https://github.com/nickjazz/trapic)
+- [Issues](https://github.com/nickjazz/trapic/issues)