Question 1

Which local models should I run, and how does hardware affect SecureContext?

Accepted Answer

SecureContext degrades gracefully by hardware tier — every LLM-powered layer fails closed (the feature quietly contributes nothing) rather than breaking ingest or search. What changes with hardware is how much of the intelligence stack is active: | Tier | Hardware | Models that fit | What you get | |---|---|---|---| | Minimum | Any CPU, ~2 GB RAM free | nomic-embed-text (embeddings only) | Hybrid BM25+vector search, working memory, audit chain, skills gating — the core. LLM layers (event extraction, entity extraction, contradiction adjudication) stay dormant. | | Mid | 8–16 GB GPU (or Apple Silicon 16 GB+) | + qwen2.5-coder:14b or phi4:14b (one at a time) | + Event-fact extraction at ingest (temporal reasoning), LLM contradiction adjudication, entity extraction, L0/L1 semantic file summaries. | | Full | 24 GB+ GPU (e.g. RTX 4090/5090) | + phi4:14b and gpt-oss:20b resident together | Everything, concurrently, at interactive latency — plus a strong local generator for QA/benchmarks. This is the configuration our published benchmark deltas were measured on. | Model-choice guidance (all measured, see bench/): Embeddings: nomic-embed-text — required, tiny, runs anywhere. Event extraction (ZCEVENTEXTRACTMODEL, default phi4:14b): our bakeoff scored phi4:14b at 100% event recall / 100% date accuracy, tying gpt-oss:20b at 2× the speed. On smaller GPUs qwen2.5-coder:14b is close behind (84.6% recall). Coder models ≠ better: qwen2.5-coder:32b scored worst (69.2%) despite being the large

Question 2

¿Qué necesito para ejecutarlo?

Accepted Answer

Node 20+ y (recomendado) Docker para el stack empaquetado de PostgreSQL + Ollama. El instalador de un solo comando hace todo en aproximadamente cinco minutos. Una alternativa SQLite funciona sin infraestructura.

Question 3

Can multiple Claude Code sessions work on the same project without conflicts?

Accepted Answer

Yes — parallel sessions atomically claim tasks from a work-stealing queue (zero double-claims at 50 agents × 100 tasks in testing), coordinate through typed broadcasts (ASSIGN/STATUS/MERGE/REJECT), and keep private per-agent memory namespaces plus a shared pool. Department-style hierarchies (heads + workers, N-tier escalation) are supported for larger agent teams.

Question 4

Are Claude Code skills safe to install?

Accepted Answer

Filesystem skills bundle scripts that run with your permissions, and Claude Code's native loader does not scan them. SecureContext adds the missing gate: AST scan at admission, HMAC verification before every execution, automatic quarantine on failure or post-admission change, and a verifiable chained log of every admission decision.

Question 5

Does SecureContext send my code or data to the cloud?

Accepted Answer

No. Memory, embeddings (Ollama nomic-embed-text), search, summarization, and the audit chain all run locally. It works fully offline (search degrades gracefully to keyword-only if Ollama is down) and costs $0 when idle.

Question 6

Can it manage a long, multi-week project?

Accepted Answer

Yes — that's what the delivery layer is for. Define a program with phases (zc_program), and the open phase's acceptance checklist travels with the project state. Closing a phase requires an evidence table and generates a checkpoint document automatically. If the orchestrating agent dies mid-phase, a fresh one reconstructs the full delivery state from two calls — verified live.

Question 7

How much does SecureContext reduce Claude Code token costs?

Accepted Answer

Measured across a 10-session project: ~87% less context-overhead (session restore is ~1.5–4k tokens instead of 20–50k of re-pasted files). On long-lived projects with hundreds of accumulated memories, the recall budget keeps each recall at ~4k tokens where an unbounded dump would be ~47k. File questions are answered from 15–50× compressed semantic summaries instead of raw reads.

Question 8

How do I give Claude Code persistent memory across sessions?

Accepted Answer

Install SecureContext (one command — see Quick start), then agents call zcremember to persist decisions and gotchas and zcrecall_context({focus: "current task"}) at session start to restore them — ranked by relevance, with natural-language time queries answered directly from memory. Memory survives restarts, compactions, and reboots because it lives in PostgreSQL, not in the context window.

Question 9

How it's different from competitors

Accepted Answer

Most "memory for Claude Code" projects do one of three things: Wrap the conversation in JSON. Saves the messages, replays them. No semantic compression, no audit trail, no security model. Sync to a cloud service. Better UX but now your agent's tool history (potentially including secrets it touched) lives on someone else's server. Drop in vector search. Adds RAG over your docs but doesn't help with the actual problem — orchestrating what the agent does, verifying what ran, surviving session compaction. SecureContext is none of those: | | SecureContext | Cloud memory services | Vector RAG plugins | Native Claude Code | |---|---|---|---|---| | Persistent across restarts | ✅ Yes, local Postgres | ✅ But on their server | ⚠️ Only docs, not state | ❌ Lossy auto-compact | | Delivery programs + evidence-gated checkpoints | ✅ zcprogram | ❌ | ❌ | ❌ | | HMAC audit trail | ✅ Per-agent subkey | ❌ | ❌ | ❌ | | Per-agent identity isolation | ✅ HKDF + RLS | ❌ | ❌ | ❌ | | Skill admission gate (HMAC + AST) | ✅ | ❌ | ❌ | ❌ | | Tamper-evident chain + session replay | ✅ | ❌ | ❌ | ❌ | | Work-stealing queue for parallel sessions | ✅ FOR UPDATE SKIP LOCKED | ❌ | ❌ | ❌ | | Runs offline | ✅ Local Postgres + Ollama | ❌ | ⚠️ Depends | ✅ | | Cost when idle | $0 | Monthly subscription | $0 | $0 | Versus the well-known memory products: Mem0 and Zep are excellent hosted memory APIs for LLM apps you build; Letta (MemGPT) is an agent framework with self-editing memory. SecureContext targets a different job — me

Question 10

What's new (v0.46.1)

Accepted Answer

Temporal retrieval round 2 + the delivery-tool kit. Compound temporal questions decompose into event clauses with rank fusion; temporal searches render a chronological Timeline block and stale results carry age warnings. zcprogram adds multi-phase delivery programs with evidence-gated, auto-checkpointing phase closure — live-verified with an orchestrator handoff test (kill the orchestrator mid-phase; a cold replacement resumes from two calls). The embed lane now has a watchdog with budgeted self-healing. Recent releases: v0.46.0 — team memory (per-user keys, shared workspaces, write attribution), session replay with cryptographic provenance, OTel Gen-AI spans, CI/headless memory CLI, chunked embeddings, compliance report export. v0.44.0 — temporal fact supersession, durable PG task graph, first public LongMemEval numbers. v0.43.0 — token-budgeted recall digest (fixes the 47k-token recall firehose on mature projects). v0.41–0.42** — memory-quality program: focused recall, consolidation, bi-temporal windows, multi-hop graph retrieval, TTL memories, multimodal ingestion. Full details in CHANGELOG.md; architecture notes in ARCHITECTURE.md.

Question 11

What is SecureContext?

Accepted Answer

Si usas Claude Code en proyectos reales, te has topado con estos obstáculos: 1. Cada sesión empieza en frío. Vuelves a pegar los mismos archivos, a volver a explicar las mismas decisiones, a redescubrir los mismos inconvenientes, y pagas tokens por el mismo trabajo de orientación una y otra vez. 2. Los proyectos largos no tienen un estado de entrega. Cuando un agente orquestador muere a mitad de una característica (crash, ventana cerrada, contexto agotado), su reemplazo comienza a ciegas. La gente compensa con archivos CHECKPOINT escritos a mano. 3. No hay registro de lo que realmente hicieron los agentes. Cuando algo falla, Claude Code no tiene un historial a prueba de manipulaciones de lo que se ejecutó. 4. Los scripts de habilidades se ejecutan con tus permisos, sin escanear. Cualquiera (o cualquier agente) con acceso de escritura a ~/.claude/skills/ puede introducir código que se ejecuta en la siguiente invocación. 5. Las sesiones paralelas se pisan entre sí. Dos ventanas reclaman la misma tarea; una sobrescribe las ediciones de la otra. SecureContext es un plugin MCP que cierra los cinco problemas. Instálalo una sola vez y tus agentes obtienen: Memoria que sobrevive a reinicios — decisiones, inconvenientes y estado del proyecto restaurados en una llamada; resúmenes semánticos por archivo que responden “qué hace este archivo” con una compresión de 15–50× en lugar de lecturas crudas; búsqueda temporal que sabe cuándo ocurrieron las cosas y marca conocimientos obsoletos. Una capa de entrega para proyectos de larga duración — zcprogram rastrea programas de múltiples fases; cerrar una fase requiere evidencia de aceptación y genera automáticamente un punto de control buscable; un orquestador nuevo reanuda el trabajo del muerto desde dos llamadas. Un registro de auditoría criptográfico — cada

designer-cognitive-instincts (Grade A)Habilidad de data-ai con pruebas de seguridad para Claude AI. Calificación A. COGNITIVE INSTINCTS — extraído de deepPrompt roles.json para diseñador

Resumen

Funciones clave

Precio

Casos de uso

Revisión de Diseño

Educación en Diseño

Pros y contras

Ventajas

Contras

Reseñas

Preguntas y respuestas

Hacer una pregunta

Alternativas a Habilidades