{"data":{"slug":"dcostenco-prism-coder","name":"prism-coder","tagline":"Persistent memory + local AI for coding agents. 2B–27B open-weight LLM fleet, cross-session Mind Palace, cognitive routing, L3 grounding verifier, multi-agent Hivemind. Works with Claude Code, Cursor,","github_url":"https://github.com/dcostenco/prism-coder","owner":"dcostenco","repo":"prism-coder","owner_avatar_url":"https://avatars.githubusercontent.com/u/64438937?v=4","primary_language":"TypeScript","stars":154,"forks":24,"topics":["agent-memory","ai-agent","anti-sycophancy","cognitive-architecture","hebbian-learning","hipaa","local-first","local-llm","mcp","mcp-server","mind-palace","model-context-protocol","ollama","prism-coder","semantic-search","spreading-activation","sqlite","supabase","typescript","vector-search"],"archived":false,"github_pushed_at":"2026-07-14T12:51:03+00:00","maintenance_label":"Very active","url":"https://www.graphcanon.com/tools/dcostenco-prism-coder","markdown_url":"https://www.graphcanon.com/tools/dcostenco-prism-coder.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/dcostenco-prism-coder","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=dcostenco-prism-coder","description":"Persistent memory + local AI for coding agents. 2B–27B open-weight LLM fleet, cross-session Mind Palace, cognitive routing, L3 grounding verifier, multi-agent Hivemind. Works with Claude Code, Cursor, VS Code. Offline-first, HIPAA-ready. Free tier included.","homepage_url":"https://synalux.ai/prism-mcp","license":"Apache-2.0","open_issues":0,"watchers":1,"ai_summary":null,"readme_excerpt":"### License: AGPL-3.0 → Apache-2.0\nPrism MCP is now Apache-2.0. The thin-client architecture means all proprietary value (skill resolution, tier gating, billing, cloud inference) lives server-side — the open client carries no moat to protect. Apache-2.0 removes the enterprise adoption friction that AGPL caused.\n\n---\n\n### Pricing — flat-rate, not per-seat\n\n| | **Prism Coder** | GitHub Copilot | Cursor | Amazon Q |\n|---|:---:|:---:|:---:|:---:|\n| **Individual** | **$19/mo** | $10/mo | $20/mo | $19/mo |\n| **Team (5 devs)** | **$49/mo flat** | $95/mo | $200/mo | $95/mo |\n| **Enterprise (25 devs)** | **$99/mo flat** | $195/mo | $1,000/mo | Custom |\n\n---\n\n---\n\n## Plans\n\nAll on-device models are free to run locally via Ollama on every tier. A subscription gates **cloud** features, higher model ceilings, and increased limits. Local model ceilings are advisory — on-device models run on your Ollama regardless of plan; the ceiling gates cloud inference and `prism_infer` routing.\n\n| | **Free** | **Standard** $19/mo | **Advanced** $49/mo | **Enterprise** $99/mo |\n|---|---|---|---|---|\n| Seats | 1 | 1 | up to 5 | up to 25 |\n| Local model ceiling | up to 4b | up to 9b | up to 27b | up to 27b |\n| Cloud inference | -- | ✅ | ✅ | ✅ (priority) |\n| Cloud Coder (Web IDE) | -- | ✅ | ✅ | ✅ (priority) |\n| Cloud search | -- | ✅ | ✅ | ✅ |\n| Max output tokens | 512 | 1,024 | 2,048 | 4,096 |\n| Cloud fallback | -- | Claude Opus 4.7 | Claude Opus 4.7 | Priority + Opus 4.7 |\n| Grounding verifier (fact-check AI output) | -- | ✅ | ✅ | ✅ |\n| Memory sync (cloud) | -- | ✅ | ✅ | ✅ |\n| Knowledge / session memory | limited | unlimited | unlimited | unlimited |\n| Analytics dashboard | -- | ✅ | ✅ | ✅ |\n| HIPAA BAA | -- | -- | -- | ✅ |\n\n14-day free trial on paid plans. 25+ seats: [contact sales](https://synalux.ai/support)\n\n---\n\n---\n\n# Install in all repos (one-time)\nbash hooks/install.sh\n\n---\n\n# Or install manually in a single repo\ncp hooks/pre-commit .git/hooks/pre-commit && chmod +x .git/hooks/pre-commit\ncp hooks/pre-push .git/hooks/pre-push && chmod +x .git/hooks/pre-push\n```\n\n| Hook | What it checks | Mode |\n|------|----------------|------|\n| `pre-commit` | Dead code, orphan services, scaffold code, missing auth | `PRECOMMIT_MODE=advisory\\|block\\|off` |\n| `pre-push` | 19-rule security audit (SSRF, SQL injection, secrets, IDOR, etc.) | `PREPUSH_MODE=advisory\\|block\\|off` |\n\nDefault mode is `advisory` (warn but allow). Set `*_MODE=block` for hard enforcement. Hooks look for full audit scripts in the repo first (`hooks/lib/`), then `~/.claude/hooks/` fallback, then minimal inline checks.\n\n---\n\n---\n\n## Self-hosting (Enterprise)\n\nRun the full model stack on your own hardware — no cloud, full data sovereignty.\n\n**Requirements:** Mac M2 Pro+ (48 GB recommended) or Linux + NVIDIA GPU, plus [Ollama](https://ollama.com).\n\n```bash\nollama pull dcostenco/prism-coder:9b       # default router\nexport LOCAL_LLM_URL=http://localhost:11434\n```\n\nRouting is automatic: `9b → 4b → cloud fallback` on desktop/server, `2b → cloud fallback` on mobile/iPhone. For iOS or another machine on the same network, run `OLLAMA_HOST=0.0.0.0 ollama serve` and point `LOCAL_LLM_URL` at the host's IP.\n\n---\n\n---\n\n## License & Tiers\n\n**This repository (the Prism MCP client)** is licensed under [Apache-2.0](./LICENSE).\n\n---\n\n### Paid (Synalux subscription)\n\nEverything in Free, plus:\n\n| Feature | Details |\n|---------|---------|\n| Model ceiling | Up to 27B locally + cloud cascade (9B → 27B → Claude) when local is unavailable |\n| Skill routing | Portal resolves which skills to load based on your project and prompt |\n| Cross-device memory | Supabase cloud sync — sessions survive across machines |\n| Grounding verifier | L3 NLI verification on model outputs |\n| Team features | Multi-agent Hivemind, workspace collaboration |\n\nThe paid tier adds **intelligent routing** — the Synalux portal determines which skills are relevant to your current project and prompt, so your agent gets domain expertise (stripe patterns,","github_created_at":"2026-02-12T21:29:53+00:00","created_at":"2026-07-15T11:01:30.660592+00:00","updated_at":"2026-07-15T11:01:34.29375+00:00","categories":[{"slug":"ai-agents","name":"AI Agents","url":"https://www.graphcanon.com/categories/ai-agents","markdown_url":"https://www.graphcanon.com/categories/ai-agents.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/ai-agents"},{"slug":"llm-frameworks","name":"LLM Frameworks","url":"https://www.graphcanon.com/categories/llm-frameworks","markdown_url":"https://www.graphcanon.com/categories/llm-frameworks.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/llm-frameworks"},{"slug":"vector-databases","name":"Vector Databases","url":"https://www.graphcanon.com/categories/vector-databases","markdown_url":"https://www.graphcanon.com/categories/vector-databases.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/vector-databases"}],"tags":[{"slug":"agent-memory","name":"agent-memory"},{"slug":"ai-agent","name":"ai-agent"},{"slug":"anti-sycophancy","name":"anti-sycophancy"},{"slug":"cognitive-architecture","name":"cognitive-architecture"},{"slug":"hebbian-learning","name":"hebbian-learning"},{"slug":"hipaa","name":"hipaa"},{"slug":"local-first","name":"local-first"},{"slug":"local-llm","name":"local-llm"}],"trust":{"provenance":{"is_fork":false,"github_id":1156623779,"owner_type":"User","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-07-15T11:01:31.651Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":14,"days_since_push":0,"last_release_at":"2026-07-11T22:44:50Z"},"security_summary":{"status":"findings","scanner":"mcp_manifest@v1","low_count":0,"high_count":0,"last_scan_at":"2026-07-15T11:01:32.100Z","medium_count":1,"scan_profile":"mcp_manifest","critical_count":0}},"capability_facts":{"mcp":{"source":"package.json:@modelcontextprotocol/*","observed_at":"2026-07-15T11:01:31.406Z","server_manifest":true},"scan":{"source":"repo_scan","observed_at":"2026-07-15T11:01:31.406Z"},"deploy":{"source":"dockerfile:Dockerfile","self_host":true,"observed_at":"2026-07-15T11:01:31.406Z","managed_saas":false},"has_cli":{"value":true,"source":"package.json:bin|scripts","observed_at":"2026-07-15T11:01:31.406Z"},"languages":{"value":["typescript","javascript"],"source":"github.language+package.json","observed_at":"2026-07-15T11:01:31.406Z"},"has_docker":{"value":true,"source":"dockerfile:Dockerfile","observed_at":"2026-07-15T11:01:31.406Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-07-15T11:01:31.406Z"}}}}