Menú
Access persistent memory across Claude sessions. Search past conversations, recall decisions, store key insights. Use when needing context from previous work or to save important information for future sessions.
Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.
Basado en la clasificación ocupacional SOC
Manage message bridge integrations (Telegram, Discord, iMessage, Google Suite, Email). List, enable, disable, configure, send messages, and look up contacts.
Manage Cognova projects - create, list, update projects. ALWAYS searches for existing projects before creating new ones. ALWAYS confirms with user before creating, showing potential matches.
Assists users in creating new Claude Code skills. Uses web search to find latest conventions and suggests alternative approaches (existing MCPs, plugins) when appropriate.
Manage Cognova tasks - create, list, update, complete tasks. Use when the user wants to track work items, todos, or action items. Can search for and associate tasks with projects.
Check system status, view configuration, troubleshoot issues with the Cognova installation. Use when diagnosing problems, checking health, or understanding the current setup.
Manage encrypted secrets for API keys, tokens, and credentials. Use when the user needs to store, retrieve, list, or delete sensitive values.
| name | memory |
| description | Access persistent memory across Claude sessions. Search past conversations, recall decisions, store key insights. Use when needing context from previous work or to save important information for future sessions. |
| allowed-tools | Bash, Read |
| metadata | {"version":"1.0.0","requires-secrets":[],"author":"Cognova","repository":"","installed-from":""} |
Access and manage persistent memory from previous Claude sessions. Memories are automatically extracted from conversations and can be explicitly stored.
| Type | Icon | Description |
|---|---|---|
decision | [D] | Architectural or design decisions made |
fact | [F] | Key facts or information learned |
solution | [S] | Solutions to problems encountered |
pattern | [P] | Patterns or conventions discovered |
preference | [*] | User or project preferences |
summary | [~] | Summarized context from sessions |
python3 ~/.claude/skills/memory/memory.py search "<query>" [options]
Options:
--type <type> - Filter by memory type--project <path> - Filter by project path--limit <n> - Max results (default: 10)Examples:
python3 ~/.claude/skills/memory/memory.py search "authentication"
python3 ~/.claude/skills/memory/memory.py search "database" --type decision
python3 ~/.claude/skills/memory/memory.py search "API" --project "/Users/me/project"
python3 ~/.claude/skills/memory/memory.py recent [limit] [options]
Options:
--type <type> - Filter by memory typeExamples:
python3 ~/.claude/skills/memory/memory.py recent
python3 ~/.claude/skills/memory/memory.py recent 5
python3 ~/.claude/skills/memory/memory.py recent --type decision
python3 ~/.claude/skills/memory/memory.py store "<content>" [options]
Options:
--type <type> - Memory type (default: fact)--project <path> - Associated project path--relevance <0-1> - Relevance score (default: 0.9)Examples:
python3 ~/.claude/skills/memory/memory.py store "We use PostgreSQL with Drizzle ORM"
python3 ~/.claude/skills/memory/memory.py store "API rate limit set to 100 req/min" --type decision
python3 ~/.claude/skills/memory/memory.py store "User prefers pnpm over npm" --type preference
python3 ~/.claude/skills/memory/memory.py decisions [options]
Options:
--project <path> - Filter by project--limit <n> - Max results (default: 20)Examples:
python3 ~/.claude/skills/memory/memory.py decisions
python3 ~/.claude/skills/memory/memory.py decisions --project "cognova"
python3 ~/.claude/skills/memory/memory.py about "<topic>" [options]
Options:
--limit <n> - Max results (default: 15)Examples:
python3 ~/.claude/skills/memory/memory.py about "error handling"
python3 ~/.claude/skills/memory/memory.py about "testing strategy"
python3 ~/.claude/skills/memory/memory.py context [options]
Shows what context would be injected into a new session.
Options:
--project <path> - Project path--limit <n> - Max memories (default: 5)When users say things like:
search or decisionsstoreaboutrecent or contextstoreMemories are automatically extracted from conversations:
When a new session starts, relevant memories are automatically injected based on:
Each memory has a relevance score (0-1):
store --type decisionabout to check relevant historydecisions to see past architectural choices