Our review
Ingests project documentation and code into a persistent semantic memory using Voyage embeddings and Qdrant vector database for cross-session retrieval.
Strengths
- Persistent memory across sessions
- Dual indexing (semantic + temporal)
- Intelligent chunking with overlap
- High recall accuracy (Recall@10 >0.98)
Limitations
- Requires Qdrant running locally
- Requires Voyage API key
- Only supports specific file types
- Memory usage of 1.4GB RAM per 1M vectors
When you need to retain context across sessions or search large project knowledge efficiently.
When the project is small and session context suffices, or you cannot set up Qdrant and an API key.
Security analysis
SafeThe skill runs local bun commands (ingest/search) on a predefined project, no destructive actions, no external data exfiltration, and no obfuscated payloads. The setup requires a local Qdrant instance and API key, but these are not transmitted elsewhere. The commands only interact with local file system and local Qdrant, posing no security risk.
No concerns found
Examples
Ingest the architecture docs from the project into Atlas so I can query them later. Use the atlas ingest command on the docs/ folder.Search Atlas memory for all mentions of 'database schema' with a limit of 10 results.Remember this decision: we will use PostgreSQL with Prisma as the ORM. Store it in Atlas for future reference.name: atlas description: Ingest project documentation and code into persistent semantic memory (Qdrant + Voyage embeddings). Use when user wants to remember context across sessions, ingest docs, or search previous work. Requires Qdrant running locally and VOYAGE_API_KEY set. allowed-tools: Bash(bun:*)
Atlas - Persistent Semantic Memory
Atlas provides automatic context ingestion and retrieval using Voyage embeddings + Qdrant vector database. Solves the context overflow problem by storing knowledge persistently across sessions.
Quick Start
Prerequisites
- Qdrant running locally:
docker run -d -p 6333:6333 qdrant/qdrant
- VOYAGE_API_KEY set (get from https://voyageai.com):
export VOYAGE_API_KEY="your-key-here"
- Verify setup:
curl http://localhost:6333/health
Ingesting Context
Store files in Atlas memory for persistent retrieval:
Ingest Single File
cd ~/production/atlas
bun run --filter @inherent.design/atlas atlas ingest /path/to/file.md
Ingest Directory (Recursive)
cd ~/production/atlas
bun run --filter @inherent.design/atlas atlas ingest /path/to/docs/ --recursive
Ingest Multiple Paths
cd ~/production/atlas
bun run --filter @inherent.design/atlas atlas ingest README.md src/index.ts docs/ -r
What gets ingested:
- Supported:
.md,.ts,.tsx,.js,.jsx,.json,.yaml,.qntm,.rs,.go,.py,.sh,.css,.html - Ignored:
node_modules,.git,dist,build,coverage,.atlas
Processing:
- Chunks text (768 tokens, 13% overlap) for semantic coherence
- Embeds with Voyage-3-large (1024-dim)
- Stores in Qdrant with dual-indexing (semantic QNTM keys + temporal timestamps)
- Preserves original text for future consolidation
Searching Context
Retrieve relevant context semantically:
Basic Search
cd ~/production/atlas
bun run --filter @inherent.design/atlas atlas search "typescript patterns"
Limited Results
cd ~/production/atlas
bun run --filter @inherent.design/atlas atlas search "memory consolidation" --limit 10
Temporal Filtering (Since Date)
cd ~/production/atlas
bun run --filter @inherent.design/atlas atlas search "sleep patterns" --since "2025-12-25"
Chronological Timeline
cd ~/production/atlas
bun run --filter @inherent.design/atlas atlas timeline --since "2025-12-01"
When to Use This Skill
Use Atlas when:
- User asks to "remember this across sessions"
- Project context is too large for single session
- User wants to search previous work/decisions
- Documentation needs to be queryable
- Building on previous research or code
Examples:
- "Remember the API architecture we discussed"
- "What did we decide about the database schema?"
- "Find all mentions of authentication patterns"
- "Ingest all the .atlas research files"
Architecture
Built on .atlas research (Steps 1-4 + Sleep Patterns):
Stack:
- Voyage-3-large embeddings (1024-dim, 9.74% better than OpenAI)
- Qdrant HNSW index (M=16, int8 quantization, 4x compression)
- RecursiveCharacterTextSplitter (semantic boundaries)
- Dual-indexing (QNTM semantic keys + RFC 3339 timestamps)
Production Config (from Step 3 research):
- Recall@10: >0.98
- Latency: 10-50ms (p95)
- Memory: 1.4GB RAM + 5GB disk per 1M vectors
Technical Details
For implementation details, see:
- docs/architecture.md - Complete technical architecture
Packages:
@inherent.design/atlas-core- Core library (embeddings, storage, search)@inherent.design/atlas- Command-line interface@inherent.design/atlas-mcp- MCP server for Claude Code integration
API Documentation Generator
Documentation
Automatically generates OpenAPI/Swagger API documentation.
Technical Writer
Documentation
Writes clear technical documentation following top style guides.
Typed Documentation Forms System
Documentation
Add typed comments, documentation, todos, and metadata to Scheme code using `(doc ...)` forms. Doc annotations are authoritative for type inference, extracted by search commands (lf-todo, lf-types), and integrated with the type checker and LSP. Useful for annotating functions, marking deprecations, or tracking localized improvements alongside the code.