Daily AI Implementation Scout Council

Today's ranked top 20

#1

newbuild nowBoth runtimes33 / 35

thedotmack/claude-mem plugin

Persistent context compression system for agent sessions, v13.4.0, Apache 2.0, Node 20+. Captures what an agent does during a Claude Code session, compresses it with AI into a structured memory block, and injects relevant context back into future sessions automatically. Install with one command: `npx claude-mem install`. Also installable via the Claude Code plugin marketplace. Works with Claude Code, Hermes, Codex, Gemini CLI, Copilot, OpenCode, and more. Listed in awesome-claude-code.

What it does for you: AXION sessions lose context between Claude Code runs and Hermes runs, forcing YY to write manual handoff documents. claude-mem automates that handoff: after each session it compresses what happened into a dated memory block, and the next session in any supported CLI reads it before starting. The AI compression step keeps the injected context within token budget even across many sessions. Hermes is explicitly listed as a supported target, which closes the most common AXION workflow gap today. Not currently in YY's enabledPlugins.

In practice: 91,362 GitHub stars, 7,992 forks, 111 stars on 2026-08-21. Number-one trending JavaScript repo today. v13.4.0 signals 13 major release cycles and sustained iteration rather than a fresh launch. Listed in awesome-claude-code with an official badge and a Trendshift ranking badge.

For: Both runtimes. Installs via npx into Claude Code as a plugin hook, but the compressed memory file it writes is vendor-agnostic markdown that Hermes and Codex read at session start. Works wherever an agent CLI reads a session context file.

Security4

Quality5

Auditability5

Useful to you5

Useful to community5

Buildable now5

Hermes4

Verdict: build now. 33 total. Same-day install via `npx claude-mem install`. New item ties mattpocock/skills at 33; new item takes rank 1 per the tiebreaker rule. Directly solves AXION's top operational gap with no infrastructure requirement beyond Node 20.

Install: run `npx claude-mem install` to hook into Claude Code automatically, or use `/plugin marketplace add thedotmack/claude-mem` then `/plugin install claude-mem` inside Claude Code. Not currently in YY's enabledPlugins. Review before enabling. Copy

Source · 91,362 GitHub stars, 7,992 forks, 111 new stars on 2026-08-21 (number-one trending JavaScript). v13.4.0, Apache 2.0, Node >=20. Mentioned in awesome-claude-code with Trendshift badge. Homepage at claude-mem.ai. First appearance in this registry.

#2

build nowBoth runtimes33 / 35

mattpocock/skills skill

Engineering skills from Matt Pocock's personal .agents directory, published as a public repo. Includes ask-matt (skill router), grill-with-docs (domain model builder updating CONTEXT.md and ADRs inline), triage (issue state machine), improve-codebase-architecture (visual HTML report of deepening opportunities), to-spec (formalises the current conversation into a spec file), to-tickets (decomposes any plan into issue-tracker tickets), and setup-matt-pocock-skills (one-time repo config). MIT license, Shell.

What it does for you: AXION engineering decisions, spec writing, and codebase architecture reviews are currently ad hoc. to-spec captures decisions already in a chat into a formal spec file; to-tickets decomposes that spec into trackable work items; improve-codebase-architecture scans any AXION component and produces a visual report. Matt Pocock uses these skills in his own daily engineering work, making them tested against real codebases rather than generic templates.

In practice: 223,647 GitHub stars, 19,244 forks. Settled from the peak of 1,214 new stars on 2026-08-20 when it was the number-one trending repo across all languages. Still in the top 25 all-languages trending on 2026-08-21. MIT license, Shell.

For: Both runtimes. SKILL.md files with no build step. Compatible with Claude Code, Hermes, and any skill-runner that reads the Agent Skills format.

Security4

Quality5

Auditability5

Useful to you5

Useful to community5

Buildable now5

Hermes4

Verdict: build now. 33 total. Second consecutive day as build now, still not adopted in AXION. Ties thedotmack/claude-mem at 33; carry loses the new-item tiebreaker. Clone and copy individual SKILL.md files into AXION/Skills today with no infrastructure requirement.

Build #2 mattpocock/skills: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/mattpocock/skills. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · 223,647 GitHub stars, 19,244 forks. MIT license, Shell. Published from the author's personal .agents directory. Last listed 2026-08-20 with verdict build now.

#3

build nowBoth runtimes32 / 35

microsoft/SkillOpt repo

Text-space optimizer from Microsoft Research that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits and validation-gated updates. Produces deployable best_skill.md artifacts. Ships SkillOpt-Sleep in v0.2.0: a nightly self-evolution engine that harvests real agent session data, mines improvement opportunities, replays and consolidates them behind a held-out validation gate. Claude Code integration shell is in the repo.

What it does for you: AXION updates skills manually. SkillOpt replaces that loop with a research-backed automated pipeline: SkillOpt-Sleep reads real Claude Code session trajectories, identifies skill weaknesses, proposes edits, validates them against held-out tests, and writes an improved best_skill.md. This is the first skill optimizer with a published paper, a PyPI release, and a Claude Code integration shell in one package.

In practice: Approximately 16,100 GitHub stars, 1,500 forks as of 2026-08-20. Microsoft Research blog feature 2026-07-24 with VentureBeat and The Decoder coverage. Third consecutive day as a build-now carry: not yet adopted in AXION.

For: Both runtimes. Python package (pip install skillopt). Ships integration shells for Claude Code, Codex, Copilot, and Devin. SKILL.md artifacts are format-agnostic and run in Hermes or any skill-compatible harness.

Security4

Quality5

Auditability5

Useful to you5

Useful to community5

Buildable now4

Hermes4

Verdict: build now. 32 total. Third consecutive day as build now, still not adopted in AXION. pip install skillopt plus the Claude Code integration shell in the repo is same-day safe.

Build #3 microsoft/SkillOpt: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/microsoft/SkillOpt. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · 16,100 stars, 1,500 forks as of 2026-08-20. v0.2.0 on PyPI released 2026-07-02. Microsoft Research blog post 2026-07-24. MIT license. Last listed 2026-08-20 with verdict build now.

#4

build nowBoth runtimes32 / 35

alirezarezvani/claude-skills skill

Community collection of 345 Claude Code and agent skills across 10 domains: engineering core (52 skills), engineering powerful (84 skills including agent-harness, zero-hallucination-coder, and skillopt-sleep), product, marketing, compliance, C-level advisory, research, business operations, commercial and finance, and daily productivity. Ships 30+ prebuilt agents and 70+ custom commands.

What it does for you: YY can clone this repo and cherry-pick from 345 battle-tested skills rather than writing from scratch. The agent-harness skill adds a goal-plan-execute-verify-close loop over any domain. The engineering/skillopt-sleep module ties this collection to the SkillOpt automated nightly improvement loop at rank 3. Updated 2026-08-18, three days before today.

In practice: Approximately 24,700 GitHub stars, 3,500 forks as of 2026-08-21. Updated 2026-08-18. Third consecutive day as a build-now carry: not yet adopted in AXION.

For: Both runtimes. Markdown skill files with no build step; compatible with Claude Code, Codex, Hermes, and any skill-runner that reads SKILL.md or CLAUDE.md format.

Security4

Quality4

Auditability4

Useful to you5

Useful to community5

Buildable now5

Hermes5

Verdict: build now. 32 total. Third consecutive day as build now. Clone and copy individual skills today.

Build #4 alirezarezvani/claude-skills: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/alirezarezvani/claude-skills. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · Approximately 24,700 GitHub stars, 3,500 forks as of 2026-08-21. Updated 2026-08-18. 345 skills across 10 domains. Last listed 2026-08-20 with verdict build now.

#5

test firstBoth runtimes31 / 35

pydantic/pydantic-ai repo

Python AI agent framework from the Pydantic team. Typed end-to-end across agents, realtime voice, image generation, and embeddings. Every major model provider supported, including Anthropic natively. Dependency injection for tools. Pydantic model validation on all LLM outputs. Supports structured streaming and multi-turn conversations with typed state.

What it does for you: AXION agents today produce untyped output requiring ad hoc JSON parsing and retry logic. Pydantic AI replaces that pattern with full typing from the LLM response through to Python objects, eliminating custom parsing code in every AXION agent. The Anthropic provider ships natively; no adapter or wrapper needed.

In practice: Approximately 19,400 GitHub stars, 2,600 forks as of 2026-08-21. MIT license. Built by the Pydantic team, whose validation library is a Python ecosystem standard. Last seen in registry 2026-08-20 with verdict test first.

For: Both runtimes. Python package; provider-agnostic typed agent framework. Native Anthropic provider ships in the package. Hermes can use it directly for typed response validation from any LLM call.

Security4

Quality5

Auditability5

Useful to you4

Useful to community4

Buildable now4

Hermes5

Verdict: test first. 31 total. The Anthropic provider is native, making this a same-session test. Try it on one AXION agent that currently does manual JSON parsing and compare output quality before committing to wider adoption.

Build #5 pydantic/pydantic-ai: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/pydantic/pydantic-ai. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · Approximately 19,400 GitHub stars, 2,600 forks as of 2026-08-21. MIT license. Anthropic provider documented at ai.pydantic.dev/api/models/anthropic/. pip install pydantic-ai. Last seen in registry 2026-08-20.

#6

test firstClaude (Axion)30 / 35

diet103/claude-code-infrastructure-showcase repo

Reference implementation of Claude Code infrastructure patterns: a skill-activation-prompt hook that fires on every user prompt and checks a skill-rules.json file for trigger patterns, suggesting relevant skills automatically. Ships a hooks README, agents README, skills README, and a Claude integration guide written for AI-assisted setup.

What it does for you: AXION already has hooks in settings.json but no systematic trigger-based skill activation. Copying the two essential hooks and one skill-rules.json from this repo introduces declarative, auditable auto-activation. With the large skill libraries at ranks 2, 3, and 4 now on the action queue, a skill-rules.json makes routing to the right skill automatic instead of manual.

In practice: Approximately 10,000 GitHub stars as of 2026-08-21. TypeScript, MIT. Last listed 2026-08-20 with verdict test first.

For: Claude (Axion). Claude Code-specific: the skill-activation-prompt hook and skill-rules.json pattern are designed for Claude Code's hook and skill runtime. Hermes does not run the same hook lifecycle.

Security4

Quality4

Auditability4

Useful to you5

Useful to community4

Buildable now5

Hermes4

Verdict: test first. 30 total. Test on an isolated project first to confirm the hook does not conflict with the build-gate.js hook already in AXION settings.json.

Build #6 diet103/claude-code-infrastructure-showcase: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/diet103/claude-code-infrastructure-showcase. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · Approximately 10,000 GitHub stars as of 2026-08-21. Updated 2026-07-13. TypeScript, MIT license. Ships CLAUDE_INTEGRATION_GUIDE.md plus three README files covering skills, hooks, and agents. Last listed 2026-08-20.

#7

test firstBoth runtimes29 / 35

akitaonrails/ai-memory repo

Rust CLI for long-term memory across agent coding sessions and cross-vendor handoff. Stores structured task context, failed approaches, open questions, and architecture notes in a vendor-agnostic format. When switching from Claude Code to another CLI in the same directory, the next agent reads the memory and continues without re-explaining the project. Ships a web UI for memory inspection and proactive query routing via Agent Skills.

What it does for you: AXION sessions switch between Claude Code and Hermes; context is currently lost unless YY writes handoffs manually. ai-memory automates that handoff via a Rust CLI. Note: thedotmack/claude-mem at rank 1 addresses the same gap via a plugin install. Both are on the test queue; ai-memory uses structured task context while claude-mem uses AI-compressed session summaries. Test both and compare recovery quality before committing to one.

In practice: 2,673 GitHub stars as of 2026-08-21. Rust 1.95+, MIT license. Last listed 2026-08-20 with verdict test first.

For: Both runtimes. Rust CLI binary that writes a vendor-agnostic memory block and Agent Skills files into any project directory. Works with Claude Code, Codex, Gemini CLI, and any agent CLI. Hermes can read the resulting memory files at session startup.

Security4

Quality4

Auditability4

Useful to you5

Useful to community5

Buildable now4

Hermes3

Verdict: test first. 29 total. Re-listed from 2026-08-20. Two cross-session memory solutions are now on the action queue: claude-mem at rank 1 and ai-memory at rank 7. Test both on a non-critical project and compare before adopting either.

Build #7 akitaonrails/ai-memory: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/akitaonrails/ai-memory. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · 2,673 GitHub stars, 231 forks, trending Rust as of 2026-08-20. Rust 1.95+, MIT license. Release badge visible in README. Last listed 2026-08-20.

#8

test firstBoth runtimes29 / 35

mukul975/Anthropic-Cybersecurity-Skills skill

817 structured cybersecurity skills for AI agents, mapped to 6 frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND, NIST AI RMF, and MITRE F3 Fight Fraud. Covers 29 security domains including threat detection, vulnerability management, incident response, and compliance. Apache 2.0. Community project; not affiliated with Anthropic despite the name.

What it does for you: AXION has no dedicated security skills today. This library adds 817 structured audit and defence skills that Claude Code can invoke when reviewing AI-generated code, checking AXION scripts for vulnerabilities, or auditing MCL platform changes. Start with the NIST CSF and D3FEND domains for pure-defence use: those skills cover asset inventory, access control review, and log analysis with no offensive techniques involved.

In practice: Approximately 29,781 GitHub stars, 3,530 forks as of 2026-08-21. Apache 2.0. Security.md and Code of Conduct included. Last listed 2026-08-20 with verdict test first.

For: Both runtimes. agentskills.io standard SKILL.md format. Works with Claude Code, GitHub Copilot, Codex CLI, Cursor, Gemini CLI, and 20+ platforms. Hermes can load SKILL.md files directly.

Security3

Quality4

Auditability4

Useful to you4

Useful to community5

Buildable now5

Hermes4

Verdict: test first. 29 total. The defence and CSF domains are immediately applicable to AXION security reviews. Test on a non-production AXION review first, picking only defence-domain skills. Do not install offensive-technique skills without explicit written authorization for the target system.

Build #8 mukul975/Anthropic-Cybersecurity-Skills: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/mukul975/Anthropic-Cybersecurity-Skills. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · Approximately 29,781 GitHub stars, 3,530 forks as of 2026-08-21. 817 skills across 29 domains. 6 framework mappings documented in README. Apache 2.0. Last listed 2026-08-20.

#9

test firstClaude (Axion)29 / 35

modelcontextprotocol/servers plugin

Anthropic-maintained collection of reference and community MCP server implementations: filesystem, GitHub, Slack, Google Drive, Postgres, fetch, memory, and others. The canonical starting point for adding MCP tools to Claude Code. No new servers added today.

What it does for you: AXION uses Claude Code with MCP tools daily. The memory server ships a knowledge graph implementation useful for AXION agent memory. The fetch server is directly relevant for the scout newsletter pipeline. Both can be installed with npx commands from this repo without a build step.

In practice: Continuously maintained by Anthropic and the MCP community. No new entries observed in today's plugin scout. Last listed 2026-08-20 with verdict test first.

For: Claude (Axion). Official MCP server reference implementations; each server is installed independently into Claude Code or any MCP client via the mcpServers block in settings.json.

Security4

Quality5

Auditability5

Useful to you3

Useful to community5

Buildable now4

Hermes3

Verdict: test first. 29 total. Re-listed from 2026-08-20. Evaluate the memory and fetch servers specifically rather than the full collection.

Install: enable individual servers by adding them to the mcpServers block in ~/.claude/settings.json following each server's README. Review before enabling. Copy

Source · Official Anthropic repository. MCP org has 24,000+ GitHub followers. Memory server and fetch server documented in the README with npx install commands. No new servers observed in today's plugin scout. Last listed 2026-08-20.

#10

test firstClaude (Axion)28 / 35

context7 (claude-plugins-official external plugin) plugin

MCP server that injects up-to-date library documentation into Claude Code context. When code references a library, context7 fetches current docs for that library version and adds them as context, preventing hallucinated API calls based on outdated training data. Listed in the Anthropic official external plugins directory.

What it does for you: AXION skills and agents frequently call Python library APIs that change between versions. context7 eliminates the class of errors where Claude generates code for an outdated API signature. Not currently in YY's enabledPlugins.

In practice: Listed in the Anthropic-managed external_plugins directory alongside GitHub, GitLab, Discord, Asana, Firebase, and fakechat. No new entries in the official plugins directory since 2026-08-20. Last listed 2026-08-20 with verdict test first.

For: Claude (Axion). MCP server installed as a Claude Code external plugin via anthropics/claude-plugins-official. Not applicable to Hermes directly.

Security4

Quality4

Auditability4

Useful to you4

Useful to community4

Buildable now5

Hermes3

Verdict: test first. 28 total. Re-listed from 2026-08-20; still not in YY's enabledPlugins. Directly installable via the official plugin directory.

Install: add context7 to enabledPlugins in ~/.claude/settings.json following the external_plugins/context7 README in anthropics/claude-plugins-official. Not currently in YY's enabledPlugins. Review before enabling. Copy

Source · Listed at github.com/anthropics/claude-plugins-official/tree/main/external_plugins/context7 as of 2026-08-21. Anthropic-reviewed. 7 external plugins in the directory as of today. Last listed 2026-08-20.

#11

test firstClaude (Axion)28 / 35

github (claude-plugins-official external plugin) plugin

Official GitHub MCP server for Claude Code. Provides direct GitHub API access from within a Claude Code session: read repos, files, issues, and PRs; create branches; commit files; open PRs. Listed in the Anthropic official external plugins directory.

What it does for you: AXION runs git operations manually via Bash tools today. The GitHub plugin lets Claude Code read issues, create branches, and open PRs directly, reducing tool-call overhead for the AXION docs repo and future tracked work. Not currently in YY's enabledPlugins.

In practice: Listed in the Anthropic-managed external_plugins directory. No new entries in the official plugins directory since 2026-08-20. Last listed 2026-08-20 with verdict test first.

For: Claude (Axion). MCP server installed as a Claude Code external plugin via anthropics/claude-plugins-official. Not applicable to Hermes directly.

Security3

Quality4

Auditability4

Useful to you4

Useful to community5

Buildable now5

Hermes3

Verdict: test first. 28 total. Re-listed from 2026-08-20; still not in YY's enabledPlugins. Test on a non-critical repo first to confirm write behaviour before enabling on the main AXION docs repo.

Install: add the github external plugin to enabledPlugins in ~/.claude/settings.json following the external_plugins/github README in anthropics/claude-plugins-official. Not currently in YY's enabledPlugins. Review before enabling. Copy

Source · Listed at github.com/anthropics/claude-plugins-official/tree/main/external_plugins/github as of 2026-08-21. Anthropic-reviewed external plugin. Last listed 2026-08-20.

#12

test firstBoth runtimes28 / 35

comet-ml/opik repo

Open-source LLM observability, evaluation, and monitoring platform from CometML. Captures multi-step agent traces: LLM calls, tool executions, retrieval steps, and agent activity in a searchable trace tree. Includes LLM-as-a-judge evaluation metrics, production dashboards, online evaluation rules, guardrails, and a PyTest integration for CI/CD evaluation on every commit.

What it does for you: AXION runs daily agent routines with no visibility into token usage, latency, prompt quality, or eval scores between runs. Opik adds an always-on trace layer that captures every AXION agent run in a searchable dashboard. YY can compare scout run quality across days, debug slow or failing nightly routines, and run automated evals on the scout newsletter output before it deploys.

In practice: Approximately 21,500 GitHub stars, 1,700 forks as of 2026-08-21. MIT license. Last listed 2026-08-20 with verdict test first.

For: Both runtimes. Python SDK plus self-hosted Docker or Opik Cloud. AXION or Hermes agents call it via a decorator or SDK wrapper. Provider-agnostic: works with any LLM.

Security4

Quality4

Auditability4

Useful to you4

Useful to community4

Buildable now4

Hermes4

Verdict: test first. 28 total. Council carry: score of 28 is at the watch floor for a repo, but all 7 axes scored 4 with no weak score and AXION has zero eval/tracing coverage today. Self-hosted Docker install is one-command. Listed as test first with this justification noted.

Build #12 comet-ml/opik: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/comet-ml/opik. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · Approximately 21,500 GitHub stars, 1,700 forks as of 2026-08-21. MIT license. pip install opik. Self-hosted via Docker or Opik Cloud. Last listed 2026-08-20.

#13

watchBoth runtimes28 / 35

microsoft/agent-framework repo

Microsoft's open framework for building, orchestrating, and deploying AI agents and multi-agent workflows in Python and .NET. Key features: multiple LLM provider support, middleware system for request/response processing, graph-based workflows with sequential/concurrent/handoff/group collaboration patterns, checkpointing, streaming, human-in-the-loop, and time-travel debugging. Blog at devblogs.microsoft.com/agent-framework/. Note: the underlying harness was previously featured on 2026-07-31 via the blog URL; this is the first listing for the GitHub repository.

What it does for you: AXION orchestrates multi-step workflows manually today, losing state on failure. agent-framework provides graph-based checkpointed workflows with human-in-the-loop as a native primitive. The Python package is pip-installable and provider-agnostic; Anthropic support needs verification against the current provider list, but the framework design is directly applicable to AXION's orchestrator-worker pattern.

In practice: 13,008 GitHub stars, 2,203 forks, 85 stars on 2026-08-21. Trending Python repo today. Microsoft organization, MIT-compatible license. Official blog at devblogs.microsoft.com/agent-framework/. First listing for this GitHub URL; underlying product previously scored build now on 2026-07-31 as a blog post.

For: Both runtimes. Python package and .NET SDK. Hermes can use the Python package directly. Provider-agnostic across multiple LLM providers.

Security4

Quality4

Auditability4

Useful to you4

Useful to community5

Buildable now3

Hermes4

Verdict: watch. 28 total. Scores 28 which is the watch level for repos (test first starts at 29). Adopting the framework requires code migration, not just installation. Watch until a concrete AXION workflow migration target is identified, then upgrade to test first.

Build #13 microsoft/agent-framework: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/microsoft/agent-framework. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · 13,008 GitHub stars, 2,203 forks, 85 stars on 2026-08-21 (trending Python). Blog at devblogs.microsoft.com/agent-framework/. Python and .NET SDK. First appearance in this registry as a GitHub URL.

#14

newwatchBoth runtimes27 / 35

PostHog/posthog repo

Open-source analytics and AI observability platform. Captures LLM traces including model, prompt, response, token cost, latency, and eval scores across every provider and framework. Ships a Self-driving feature: a scout watches cost, latency, errors, volume, and eval regressions and routes investigation reports to a human inbox when a slice regresses against its baseline. Also includes analytics, session replay, feature flags, experiments, and error tracking. MCP server for steering from Claude Code or any MCP client. Python, MIT-compatible license, self-hostable.

What it does for you: AXION runs daily agent routines with no eval or regression visibility. PostHog's AI observability layer would capture every AXION agent run in a trace tree and automatically surface regressions, quality drops, or cost spikes as inbox reports. The Self-driving scout stops at a human-routed report rather than opening PRs autonomously, which fits AXION's approval-gate model. Compare with comet-ml/opik at rank 12 before choosing an observability platform.

In practice: 37,961 GitHub stars, 3,213 forks, 100 stars on 2026-08-21 (trending Python today). MIT-compatible license. Mature platform with full docs at posthog.com. Found via lens 2 Python trending today.

For: Both runtimes. Python SDK that AXION or Hermes agents call to emit trace events. Self-hostable via Docker. MCP server ships for steering from any MCP client including Claude Code.

Security4

Quality5

Auditability4

Useful to you3

Useful to community5

Buildable now3

Hermes3

Verdict: watch. 27 total. Comprehensive platform but buildable_now scored 3 because self-hosting PostHog for AI observability alone is oversize for AXION's current need. Watch alongside opik at rank 12; upgrade to test first once AXION has a specific eval metric to track.

Build #14 PostHog/posthog: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/PostHog/posthog. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · 37,961 GitHub stars, 3,213 forks, 100 stars on 2026-08-21 (trending Python). MIT-compatible license. AI observability docs at posthog.com/docs/ai-observability. First appearance in this registry.

#15

watchBoth runtimes27 / 35

huggingface/smolagents repo

Barebones Python library for agents that plan in code. CodeAgent generates executable Python plans that run locally; ToolCallingAgent uses standard tool-call JSON. LLM-agnostic: HuggingFace inference, Anthropic via LiteLLM, OpenAI-compatible servers, and local models all supported. Agents can be pushed to and loaded from the HuggingFace Hub for cross-machine sharing.

What it does for you: smolagents is the simplest possible agent framework: no heavy abstraction, agents plan and execute as Python code. The Hub push/pull enables versioned AXION agent sharing across machines, complementing the Mac brain sync setup. Re-listed alongside thedotmack/claude-mem (rank 1) to note that claude-mem handles session memory while smolagents handles the agent execution model; the two are complementary rather than competing.

In practice: Approximately 28,900 GitHub stars, 2,900 forks as of 2026-08-21. MIT license. HuggingFace official repo. Last listed 2026-08-20 with verdict watch.

For: Both runtimes. Python package, LLM-agnostic. Anthropic Claude is supported via the LiteLLMModel wrapper. Hermes can use smolagents workflows directly from Python.

Security4

Quality4

Auditability4

Useful to you3

Useful to community5

Buildable now4

Hermes3

Verdict: watch. 27 total. Claude requires the LiteLLMModel wrapper rather than a native provider, adding a dependency layer. Watch for a native Anthropic provider or until AXION has a concrete use case for a code-planning agent.

Build #15 huggingface/smolagents: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/huggingface/smolagents. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · Approximately 28,900 GitHub stars, 2,900 forks as of 2026-08-21. MIT license. pip install smolagents. Native Anthropic path uses LiteLLM wrapper per the README. Last listed 2026-08-20.

#16

watchBoth runtimes27 / 35

langchain-ai/langgraph repo

Low-level graph-based orchestration framework for building stateful, long-running agents. Checkpointed state machine layer that survives failures and restarts. Production users include Klarna, Replit, and Elastic. The Deep Agents extension adds higher-level planning, subagent delegation, and file-system context for complex multi-step tasks.

What it does for you: AXION orchestrates multi-step workflows manually today. LangGraph provides a checkpointed state machine layer. Re-listed alongside microsoft/agent-framework at rank 13 to put both graph-based orchestration options on the radar simultaneously so YY can compare before choosing. LangGraph has more production case studies; agent-framework has the Microsoft organization backing and declarative-agents tooling.

In practice: Approximately 40,000 GitHub stars, 6,700 forks as of 2026-08-21. MIT license. Production case studies at Klarna, Replit, Elastic. Last listed 2026-08-20 with verdict watch.

For: Both runtimes. Python package; graph state management is provider-agnostic. Hermes can use LangGraph workflows directly. The Deep Agents extension is also Python.

Security3

Quality5

Auditability4

Useful to you4

Useful to community5

Buildable now3

Hermes3

Verdict: watch. 27 total. Re-listed from 2026-08-20 because microsoft/agent-framework at rank 13 introduces a directly comparable orchestration framework. Held at watch until AXION has a concrete workflow migration target.

Build #16 langchain-ai/langgraph: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/langchain-ai/langgraph. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · Approximately 40,000 GitHub stars, 6,700 forks as of 2026-08-21. MIT license. pip install -U langgraph. Production case studies for Klarna, Replit, Elastic at the LangGraph site. Last listed 2026-08-20.

#17

watchBoth runtimes27 / 35

run-llama/llama_index repo

Data framework for connecting LLMs to external data sources via indexing, retrieval, and query engines. Over-the-box connectors for documents, databases, APIs, and web pages. 100+ data source integrations. Structured loaders, vector and keyword indexing, and query pipelines with Anthropic Claude support built in.

What it does for you: AXION's brain ingest is a custom pipeline handling PDFs, web pages, and documents with ad hoc parsing. LlamaIndex would standardize that pipeline: structured loaders handle format detection, indexing provides search, and the query engine returns cited answers. The Anthropic provider means no adapter is needed. Active 30-day re-evaluation timer: if document ingestion has not been prioritised in AXION by 2026-09-20, this item moves to skip.

In practice: 40,000+ GitHub stars. MIT license. Last listed 2026-08-20 with verdict watch and 30-day timer started.

For: Both runtimes. Python package; provider-agnostic data framework. Anthropic Claude is supported directly. Hermes can use LlamaIndex loaders and query engines from Python.

Security3

Quality4

Auditability4

Useful to you3

Useful to community5

Buildable now4

Hermes4

Verdict: watch. 27 total. Re-listed from 2026-08-20 to keep the 30-day timer visible. Day 1 of 30: move to skip on 2026-09-20 if document ingestion is not prioritised.

Build #17 run-llama/llama_index: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/run-llama/llama_index. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · 40,000+ GitHub stars. MIT license. pip install llama-index. Anthropic provider documented at docs.llamaindex.ai. Last listed 2026-08-20 with verdict watch.

#18

watchBoth runtimes26 / 35

agno-agi/agno repo

Agent platform SDK with 100+ pre-built tool integrations, cron-based scheduling, Telegram and WhatsApp interfaces, human approval workflow, JWT-based RBAC, multi-tenant isolation, and OpenTelemetry tracing. Designed to build and manage production agent platforms.

What it does for you: AXION uses Telegram for YY notifications, runs cron schedules via the bridge, and needs human approval gates for destructive operations. Agno ships all three as native primitives. Re-listed 8 days after the last listing (2026-08-13) alongside microsoft/agent-framework at rank 13: both are multi-agent orchestration platforms and comparing them before committing to either is worthwhile.

In practice: Approximately 41,800 GitHub stars, 5,800 forks as of 2026-08-21. MIT license. Last listed 2026-08-13 with verdict watch.

For: Both runtimes. Python SDK with a production API, 50+ endpoints, SSE and websocket support. AXION agents call it from Python; Hermes can submit work to an Agno-managed agent via HTTP.

Security3

Quality4

Auditability4

Useful to you4

Useful to community5

Buildable now3

Hermes3

Verdict: watch. 26 total. Re-listed 8 days after last listing. Adopting Agno as the AXION runtime is a major architectural commitment. buildable_now scored 3 because adoption requires code migration, not just installation.

Build #18 agno-agi/agno: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/agno-agi/agno. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · Approximately 41,800 GitHub stars, 5,800 forks as of 2026-08-21. MIT license. pip install agno. Production API documented at docs.agno.com. Last listed 2026-08-13.

#19

newwatchClaude (Axion)24 / 35

cursor/plugins repo

Cursor's official plugin specification and listing of official Cursor plugins. Defines a multi-plugin marketplace repository structure: each plugin folder contains a .cursor-plugin/plugin.json manifest, a skills/ directory with SKILL.md files, rules/ for .mdc files, and an mcp.json for MCP server definitions. TypeScript. The official Cursor plugins published here follow this specification and are installable from the Cursor plugin marketplace.

What it does for you: AXION already uses Claude Code plugins. Studying the cursor/plugins spec shows how a major IDE builds its plugin marketplace using the same SKILL.md and MCP primitives. The official Cursor plugins listed here include MCP server definitions that could be adapted for Claude Code without rebuilding from scratch. Pattern reference for AXION's own skill and plugin authoring work.

In practice: 4,040 GitHub stars, 338 forks, 473 stars on 2026-08-21 (number-one trending TypeScript repo today). Official Cursor organization. TypeScript. First appearance in this registry.

For: Claude (Axion). Cursor-specific plugin format, but the skills (SKILL.md), rules (.mdc), and MCP server definitions (.mcp.json) in official Cursor plugins are format-compatible with Claude Code. MCP server configs can be adapted directly.

Security4

Quality4

Auditability4

Useful to you3

Useful to community4

Buildable now3

Hermes2

Verdict: watch. 24 total. Pattern reference: the plugin spec and official MCP definitions are worth studying but not directly deployable into AXION without adaptation. Watch for specific Cursor official plugins that wrap MCP servers AXION could reuse.

Build #19 cursor/plugins: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/cursor/plugins. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · 4,040 GitHub stars, 338 forks, 473 new stars on 2026-08-21 (number-one trending TypeScript). Official Cursor organization. TypeScript. First appearance in this registry.

#20

newwatchClaude (Axion)23 / 35

vercel-labs/agent-skills skill

Vercel's official collection of agent skills. Ships CLAUDE.md and AGENTS.md for cross-platform support alongside a skills/ directory with SKILL.md files. The documented vercel-optimize skill audits a Vercel project for cost, performance, reliability, caching, function usage, and billing opportunities. Additional skills in the collection follow the same Vercel deployment domain.

What it does for you: AXION does not use Vercel, so the individual skills are not directly applicable. The value is architectural: this is the reference implementation of how a major platform publishes an official agent skill collection using CLAUDE.md, AGENTS.md, skills/, and packages/ in a single repository. Study the structure before designing AXION's own skill distribution pattern.

In practice: 30,273 GitHub stars, 2,706 forks, 49 stars on 2026-08-21 (trending JavaScript today). Official Vercel labs organization. JavaScript. First appearance in this registry.

For: Claude (Axion). SKILL.md format compatible with Claude Code and other Agent Skills runtimes, but content is Vercel deployment-specific. Not applicable to Hermes in AXION's context.

Security4

Quality4

Auditability4

Useful to you2

Useful to community5

Buildable now2

Hermes2

Verdict: watch. 23 total. Vercel-specific content makes this a pattern reference for AXION, not a directly usable skill set. Watch for any vendor-neutral skills added to the collection in future releases.

Build #20 vercel-labs/agent-skills: use the ai-implementation-build-intake skill to build this safely. Source: https://github.com/vercel-labs/agent-skills. Save canonical skill/agent under AXION\Skills and AXION\Agents. Copy

Source · 30,273 GitHub stars, 2,706 forks, 49 new stars on 2026-08-21 (trending JavaScript). Official Vercel labs organization. JavaScript. CLAUDE.md and AGENTS.md present in repo root for cross-platform support. First appearance in this registry.