Best new AI-agent and productivity resources: day 33

Yesterday's directory additions were especially useful for people building with AI agents, but there are also practical picks for organizing work and creating documents. Here are the strongest starting points, grouped by what they help you do. These are descriptions of the listed projects, not guarantees of performance or security; check each project's documentation before using it for consequential work.

Keep agent work organized

Wake brings AI-agent sessions into one searchable desktop view, so developers can find and resume work without digging through terminal history. Start here if you regularly have several coding sessions going at once. https://github.com/iAmCorey/Wake

Agentsmith offers a model-agnostic harness with work-type profiles for agents such as Claude, Codex and Gemini. It is a better fit for developers standardizing repeatable agent workflows than for someone who only needs occasional chat. https://github.com/PromptPartner/agentsmith

Context Loader turns a local Git repository into bounded Markdown or JSON context. Use it when a coding assistant needs a focused snapshot of a codebase rather than an unfiltered dump. https://github.com/xuanheng-tech/context-loader

Add checkpoints before agents act

Prismor is a self-hosted control plane for observing, approving or blocking agent tool calls. Its human-approval path is worth examining when an agent can take actions that are hard to undo. https://github.com/PrismorSec/prismor

Microsoft's Agent Governance Toolkit brings policy, identity and sandboxing concepts together for teams working with autonomous agents. It is aimed at engineering and security teams, not casual users. https://github.com/microsoft/agent-governance-toolkit

Govern puts a signed decision gate before a deployment or agent workflow step; Writ applies workflow gates and relevant rules inside Claude Code. The first is for teams managing CI and deployment decisions, while the second is specific to Claude Code workflows. https://github.com/decionis/govern

Writ: https://github.com/infinri/Writ

Also added: nono for agent runtime isolation and Agent Control Standard for framework-independent runtime-control conventions. Neither removes the need to review permissions and test a workflow yourself. https://github.com/nolabs-ai/nono

Agent Control Standard: https://github.com/GenAI-Security-Project/agent-control-standard

Build reusable skills and knowledge workflows

SkillNet focuses on creating and evaluating reusable agent skills, while Skillware packages modular skills in Python. Choose SkillNet if evaluating skills matters; look at Skillware if you need a Python framework for organizing them. https://github.com/zjunlp/SkillNet

Skillware: https://github.com/ARPAHLS/skillware

Claude Skills Collection provides a broad set of skills and agents for several coding assistants. Browse selectively and review individual skills before installing them into a tool with access to your files. https://github.com/borghei/Claude-Skills

Vercel Labs' Knowledge Agent Template is a starting point for an agent grounded in a file-system knowledge base. It suits developers who want to adapt a template around their own documentation rather than adopt a finished chatbot. https://github.com/vercel-labs/knowledge-agent-template

Everyday productivity beyond coding

Bento is a portable, offline-oriented office suite for documents, spreadsheets and presentations. It is the most direct pick here for someone who wants to work on files without depending on a cloud office account. https://github.com/nyblnet/bento

Open Supermarkets combines command-line, MCP and agent skills for grocery search, comparison and checkout across supported providers. Check its country and supermarket coverage first, and review any purchase before confirming it. https://github.com/abracadabra50/open-supermarkets

More specialist additions

Sentio gives developers an email server and structured mail workflows for agents. Oracle VecDB Python SDK targets vector search and retrieval on Oracle AI Database. Sphere SDK explores identity and transactions between agents. https://github.com/truespar/sentio

Oracle VecDB Python SDK: https://github.com/oracle/vecdb-python-sdk

Sphere SDK: https://github.com/unicity-sphere/sphere-sdk

DomA explores browser automation, UniFace brings several face-analysis tasks into one Python interface, and ainovel-cli experiments with multi-agent novel creation. Pick these only when their particular workflow matches your project. https://github.com/dom-actions/doma

UniFace: https://github.com/yakhyo/uniface

ainovel-cli: https://github.com/voocel/ainovel-cli

The remaining additions include Prismor's neighboring research and discovery resources: Awesome Multi-Agent Orchestrators, Awesome TypeSafe Jev, Fable 5 Opus 5.5 Orchestrator and DSH NextTavern. Browse the entire hand-curated directory to compare the new entries with older alternatives. https://bestproductivityresources.com/directory