MCP App StoreMCP App Store
from total-recall264
Autonomous agent memory that passively observes session transcripts, compresses them into prioritized notes, consolidates redundancies, and recovers missed cont
Total Recall adds a persistent, observation-driven memory layer to an agent. It periodically scans recent session transcripts, compresses them into prioritized observations, consolidates and prunes redundant items, and provides session-recovery hooks to capture missed context. Optional reactive watching makes capture near-real-time on high activity.
Use Total Recall when you want cross-session continuity without managing databases or vector indexes: long-running agents, teams needing concise memory logs, or agents that must preserve important events with minimal maintenance. It's ideal for agents that should "just pay attention" and surface prior decisions, preferences, or follow-ups.
prompts/ for Observer/Reflector.Works with agents that can run shell scripts and call LLM providers (OpenRouter, Ollama, Groq). Particularly suited to OpenClaw-style agents and cron-driven agent runs.
Total Recall is a comprehensive autonomous agent memory skill with 17 scripts covering observation compression, reflection/consolidation, session recovery, reactive watching, dream cycle, importance decay, and emergency alerting. The architecture is well-designed with 5 layers of redundancy. Scripts that ran successfully (config, tools, buffer-inject, emergency-surface, sensor-sweep, staging-review) showed clean output with proper logging and graceful degradation. 6 scripts failed due to _compat.sh source path issues in the test environment and missing jq. The skill requires OPENROUTER_API_KEY and several system deps (jq, inotify-tools, PyYAML). Security is solid — no exfiltration, no destructive commands, API keys are env-based — with minor concerns around eval-based .env loading and some string interpolation in shell commands.
jqinotify-toolspython3-yaml (PyYAML)Impressively comprehensive skill. The 5-layer redundancy (observer, reflector, session recovery, reactive watcher, pre-compaction hook) is well-architected. Dream cycle with importance decay and pattern promotion is a standout feature. Code is generally well-structured with atomic writes, lock files, dedup, and graceful degradation. The eval-based .env sourcing and some shell interpolation patterns are minor security concerns but not malicious. Scripts that need workspace context naturally fail in isolation — expected behavior.
Development Worktree
Create an isolated git worktree for feature work, auto-run project setup, and verify a clean test baseline before development.
Readwise Reader Document Management
Manage Readwise Reader documents: list, save, search, move, tag, highlight, export and bulk-edit via official and custom CLIs.
Bounty Hunter — Atlas
Persona skill: 'Atlas' — a profit-focused developer persona for discovering, evaluating and executing paid bounties or freelance tasks with ROI-aware workflows.
Junshi — Research Advisor
Daily strategic research advisor that scans arXiv/venues, digests papers, and proposes bold, ranked research ideas tailored to the user's profile.
Full Stack Builder
End-to-end builder that scaffolds, implements, tests, and optionally deploys web and API applications from a natural-language specification.
ezBookkeeping API Tools
Command-line API tools for ezBookkeeping: record and query transactions, retrieve accounts/categories/tags, and fetch exchange rates for self-hosted personal fi
Feishu Voice Sender
Convert MP3s and send them as native Feishu voice messages (playable voice clips) to users or groups.
Claw Bench
Benchmarking skill that guides an agent through a structured suite of capability tests and reporting steps for leaderboard submission.