Agent skill

Nextclaw HTTP Agent Runtime Integration

by Peiiii in Peiiii/nextclaw

A skill your agent uses when integrating or debugging a Hermes-style external HTTP agent runtime in NextClaw, including runtime config, session selection, readiness, streaming, and adapter behavior.

MITAuto-check passedDevelopment

Install Nextclaw HTTP Agent Runtime Integration

skills CLI
$ npx skills add Peiiii/nextclaw --skill nextclaw-http-agent-runtime-integration -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install Peiiii/nextclaw nextclaw-http-agent-runtime-integration --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/Peiiii/nextclaw.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/wiki/skills/architecture/nextclaw-http-agent-runtime-integration .claude/skills/nextclaw-http-agent-runtime-integration && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
nextclaw-http-agent-runtime-integration
GitHub stars
260
Token cost
~1.9k tokens
SKILL.md length
853 words
Files
2
Skills in repo
70
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when integrating or debugging a Hermes-style external HTTP agent runtime in NextClaw, including runtime config, session selection, readiness, streaming, and adapter behavior.

  • Works in 6 steps: Add or update an… → Point config.baseUrl at the external… → Set label to the user-facing runtime… → …
  • Debugging a Hermes-style external HTTP agent runtime in NextClaw
  • SKILL.md covers Overview, When To Use, Core Rule and First Decision, plus 8 more sections
  • Calls pnpm

What it does

Nextclaw HTTP Agent Runtime Integration is an agent skill from Peiiii/nextclaw. Use when integrating or debugging a Hermes-style external HTTP agent runtime in NextClaw, including runtime config, session selection, readiness, streaming, and adapter behavior.

Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).

It sits in Development, covering Debugging. The repository describes itself as: A human-centered long-term AI partner—not a task-centered assistant. The licence is MIT.

When your agent uses it

  • Debugging a Hermes-style external HTTP agent runtime in NextClaw
  • Including runtime config
  • Session selection
  • Adapter behavior

Example prompts

  • “/nextclaw-http-agent-runtime-integration”

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. Add or update an agents.runtimes.entries. entry with type: "narp-http".
  2. Point config.baseUrl at the external runtime or adapter.
  3. Set label to the user-facing runtime name, for example Hermes.
  4. Optionally set
  5. Verify that the session type becomes visible and ready.
  6. Run a real message smoke, not only static config inspection.

What it can do on your machine

Read from SKILL.md and the folder at commit 973722e. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • pnpm

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use pnpm, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Nextclaw HTTP Agent Runtime Integration loads about 1.9k tokens when it runs. Until then it costs about 55 tokens; SKILL.md has 853 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~55
When it runs · the whole SKILL.md, loaded when a task matches
~1.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Peiiii/nextclaw at commit 973722e, republished under its MIT licence (© Peiiii). 853 words, ~1,865 tokens.

Download SKILL.mdSave it as .claude/skills/nextclaw-http-agent-runtime-integration/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
nextclaw-http-agent-runtime-integration
description
Use when integrating or debugging a Hermes-style external HTTP agent runtime in NextClaw, including runtime config, session selection, readiness, streaming, and adapter behavior.

Integrating HTTP Agent Runtime

Overview

Use this skill to connect an external agent runtime into NextClaw through the generic narp-http path.

Default architecture:

  • Keep NextClaw core generic.
  • Reuse the existing builtin narp-http runtime provider and HTTP client.
  • Treat Hermes or any other backend as an external runtime behind an HTTP adapter.
  • End with a real selectable session type in NextClaw, not just a design note.

Read these first when the task is non-trivial:

  • docs/VISION.md
  • docs/plans/2026-04-15-http-backed-agent-runtime-design.md
  • docs/logs/v0.16.28-http-backed-agent-runtime-mvp/README.md

When To Use

Use this skill when any of these are true:

  • The user wants Hermes to appear as a NextClaw session type.
  • The user wants to connect an external agent runtime by HTTP address.
  • The user wants “just configure it” runtime onboarding instead of hard-coding a new runtime.
  • The user wants the AI to debug why an HTTP runtime is not ready, not visible, not streaming, or returning errors.

Do not use this skill for:

  • MCP server integration
  • ordinary tool/plugin integration that does not create an agent runtime
  • adding a brand-new non-HTTP runtime family

Core Rule

Prefer this split unless the user explicitly asks for something else:

  • NextClaw owns the generic narp-http provider and session-type experience.
  • The external runtime owns its private protocol.
  • An adapter server translates the private protocol into the shared HTTP/SSE contract.

Do not create a Hermes-specific runtime kind if narp-http can solve it.

First Decision

Classify the task into exactly one path:

Path A: Existing HTTP runtime is already ready

Choose this when the target already exposes the expected endpoints:

  • POST /send
  • GET /stream?sessionId=...
  • POST /abort
  • optional GET /health

In this path, the AI should mostly do configuration and verification.

Path B: Adapter server is still required

Choose this when the target runtime only exposes:

  • private SDK calls
  • OpenAI-compatible APIs
  • CLI commands
  • ACP or another private wire protocol

In this path, the AI should first add or update an adapter server, then wire NextClaw to the adapter URL.

Files To Reuse

Start from the existing implementation before inventing new structure:

  • packages/nextclaw-ncp-runtime-http-client
  • packages/nextclaw-ncp-runtime-adapter-hermes-http
  • packages/nextclaw/src/cli/commands/ncp/runtime/create-ui-ncp-agent.http-runtime.test.ts
  • packages/nextclaw/src/cli/commands/ncp/runtime/create-ui-ncp-agent.hermes-http-runtime.test.ts

Important current boundary:

  • narp-http is already the generic runtime kind.
  • Hermes is already implemented as an external adapter, not as a core runtime kind.
  • Current MVP is configuration-driven for one plugin entry, not yet a full multi-instance runtime catalog.

Path A Workflow

When the external runtime already serves the HTTP contract:

  1. Add or update an agents.runtimes.entries.<id> entry with type: "narp-http".
  2. Point config.baseUrl at the external runtime or adapter.
  3. Set label to the user-facing runtime name, for example Hermes.
  4. Optionally set:
    • basePath
    • healthcheckUrl
    • recommendedModel
    • supportedModels
    • headers
  5. Verify that the session type becomes visible and ready.
  6. Run a real message smoke, not only static config inspection.

Example config:

json
{
  "agents": {
    "runtimes": {
      "entries": {
        "hermes": {
          "enabled": true,
          "label": "Hermes",
          "type": "narp-http",
          "config": {
            "baseUrl": "http://127.0.0.1:18765",
            "basePath": "/ncp/agent",
            "healthcheckUrl": "http://127.0.0.1:18765/health",
            "recommendedModel": "hermes-agent"
          }
        }
      }
    }
  }
}

Path B Workflow

When an adapter is needed:

  1. Keep NextClaw side generic.
  2. Add or update a standalone adapter package under packages/extensions/.
  3. Make the adapter expose the shared HTTP/SSE contract.
  4. Keep target-runtime-specific translation inside the adapter package.
  5. Only after the adapter is runnable, point narp-http config at its baseUrl.

For Hermes, prefer the existing adapter package instead of rebuilding another integration path.

Show full SKILL.md (340 more words)Show less

Debugging Checklist

If the runtime does not appear in session type selection:

  • check the agents.runtimes.entries.<id> config and enabled: true
  • check baseUrl
  • if using readiness probing, check healthcheckUrl
  • run a direct curl against /health

If the runtime appears but is not ready:

  • inspect healthcheck response
  • verify adapter process is actually listening
  • verify required auth headers or API keys are present

If the runtime is selectable but sending fails:

  • test POST /send directly
  • test GET /stream?sessionId=... directly
  • check whether the adapter emits run.error or only hangs
  • inspect whether the upstream provider rejected credentials

If the run finishes but text is empty:

  • inspect upstream provider logs first
  • confirm the adapter is parsing the upstream stream format correctly
  • for Hermes, check whether the upstream provider returned 401 or 403

Validation

Do not stop after editing files. Run the smallest sufficient end-to-end checks.

Prefer these validations when relevant:

bash
pnpm -C packages/nextclaw-ncp-runtime-http-client test
pnpm -C packages/nextclaw-ncp-runtime-adapter-hermes-http test
pnpm -C packages/nextclaw-ncp-runtime-adapter-hermes-http lint
pnpm -C packages/nextclaw-ncp-runtime-adapter-hermes-http tsc
pnpm -C packages/nextclaw-ncp-runtime-adapter-hermes-http build
pnpm -C packages/nextclaw test -- create-ui-ncp-agent.http-runtime.test.ts
pnpm -C packages/nextclaw test -- create-ui-ncp-agent.hermes-http-runtime.test.ts

When the user asks for real verification, use an isolated temp home instead of writing smoke data into the repo.

Real Hermes Smoke

When Hermes local source is available, the AI may use this pattern:

  1. start Hermes API server with isolated HERMES_HOME
  2. confirm /health and /v1/models
  3. start packages/extensions/nextclaw-ncp-runtime-adapter-hermes-http/dist/cli.js
  4. point NextClaw narp-http config at the adapter URL
  5. send a real message and inspect returned event types and text deltas

If the bridge works but the provider rejects the call, report that explicitly as an upstream credential or permission issue, not as a broken NextClaw integration.

Output Contract

When finishing this task, the AI should report:

  • which path it used: Path A or Path B
  • which URL NextClaw should point to
  • whether the session type is visible and ready
  • whether automated tests passed
  • whether real smoke passed
  • if real smoke failed, whether the failure is in NextClaw, adapter, or upstream provider

Common Mistakes

  • creating a Hermes-specific core runtime instead of reusing narp-http
  • treating a skill as if it can replace the runtime contract
  • wiring config but skipping real send/stream verification
  • reporting “integration complete” when only /health works
  • blaming NextClaw for upstream provider auth failures

© Peiiii, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in .agents/wiki/skills/architecture/nextclaw-http-agent-runtime-integration of Peiiii/nextclaw.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit 973722e

Compare with similar skills

Nextclaw HTTP Agent Runtime Integration next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Nextclaw HTTP Agent Runtime Integration compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Nextclaw HTTP Agent Runtime Integration this skillPeiiii/nextclaw260—~1.9kAutomated safety check: PassMIT
Trellis Session Insightmindfold-ai/Trellis15k4 repos~1.7kAutomated safety check: PassAGPL-3.0
Native Data FetchingCherryHQ/cherry-studio-app4k6 repos~2.9kAutomated safety check: NotesMIT
Debugging Executionsn8n-io/n8n207k—~2.6kAutomated safety check: PassCustom licence
Aoti Debugpytorch/pytorch104k1 repos~1.7kAutomated safety check: PassCustom licence
Herdr Throwaway Reproductionherdrdev/herdr43k—~2.4kAutomated safety check: PassApache-2.0

Similar skills

  • Trellis Session Insight

    mindfold-ai/Trellis

    Reach into past AI conversation history through the trellis mem CLI.

    15k GitHub starsUsed in 4 repos~1.7k tokens
    DevelopmentAuto-check passed
  • Native Data Fetching

    CherryHQ/cherry-studio-app

    A skill your agent uses when implementing or debugging ANY network request, API call, or data fetching.

    4k GitHub starsUsed in 6 repos~2.9k tokens
    DevelopmentAuto-check: notes
  • Official

    Debug failed or wrong-output workflow executions using executions tools.

    207k GitHub stars~2.6k tokensUpdated today
    DevelopmentAuto-check passed
  • Aoti Debug

    pytorch/pytorch

    Debug AOTInductor (AOTI) errors and crashes. An agent skill from pytorch/pytorch.

    104k GitHub starsUsed in 1 repo~1.7k tokens
    DevelopmentAuto-check passed
  • Runs a disposable, uniquely named Herdr session inside an existing one so runtime, pane, terminal or API bugs can be reproduced without touching the main session.

    43k GitHub stars~2.4k tokensUpdated today
    DevelopmentAuto-check passed
  • Systematic Debugging

    ultralisp/ultralisp

    A skill your agent uses when encountering any bug, test failure, or unexpected behavior, before proposing fixes

    258 GitHub starsUsed in 51 repos~2.4k tokens
    DevelopmentAuto-check passed

More from Peiiii/nextclaw

All 70 skills in this repo
  • Nextclaw Skin Studio

    Peiiii/nextclaw

    A skill your agent uses when a user wants to browse, apply, inspect, create, refine, switch, or remove a NextClaw skin; wants a personal skin with arbitrary CSS, JavaScript, images, SVG, DOM…

    260 GitHub stars~1.8k tokensUpdated yesterday
    Auto-check passed
  • A skill your agent uses when a development task needs visible phase tracing, task-level or phase-level Token measurement, model/effort comparison, or a deterministic usage report or local dashboard…

    260 GitHub stars~802 tokensUpdated yesterday
    Auto-check passed
  • 当 NextClaw 产品更新后需要生成、替换、挑毛病或检查官网、GitHub README、用户文档或社交传播中的真实截图、AI 宣传视觉、整页 HTML 宣传预览、社区二维码等对外视觉资产时使用;也用于“更新截图”“重新截一批图”“做宣传页”“生成 campaign 页面”“视觉审稿”“五星挑刺法”或发布前检查视觉资产。普通站点布局开发或只写文章不触发。

    260 GitHub stars~801 tokensUpdated yesterday
    Auto-check passed
  • UI UX Pro Max

    Peiiii/nextclaw

    A skill your agent uses when the user wants professional UI/UX design guidance, design-system generation, UX review, or stack-specific frontend guidance through a bundled local UI/UX Pro Max dataset…

    260 GitHub stars~2k tokensUpdated yesterday
    Auto-check passed
  • Impeccable

    Peiiii/nextclaw

    A skill your agent uses when the user wants distinctive, production-grade frontend design, anti-generic AI aesthetics, UX critique, technical UI audits, or final polish through bundled Impeccable…

    260 GitHub stars~2.5k tokensUpdated yesterday
    Auto-check passed
  • Curate NextClaw skill resources, including OpenClaw and community sources.

    260 GitHub stars~479 tokensUpdated yesterday
    Auto-check passed

Categories

Questions about Nextclaw HTTP Agent Runtime Integration

What does Nextclaw HTTP Agent Runtime Integration do?

A skill your agent uses when integrating or debugging a Hermes-style external HTTP agent runtime in NextClaw, including runtime config, session selection, readiness, streaming, and adapter behavior. Nextclaw HTTP Agent Runtime Integration is an agent skill from Peiiii/nextclaw. Use when integrating or debugging a Hermes-style external HTTP agent runtime in NextClaw, including runtime config, session selection, readiness, streaming, and adapter behavior.

When should I use Nextclaw HTTP Agent Runtime Integration?

Nextclaw HTTP Agent Runtime Integration fits situations like: debugging a Hermes-style external HTTP agent runtime in NextClaw; including runtime config; session selection; adapter behavior.

How do I install Nextclaw HTTP Agent Runtime Integration in Claude Code?

Run `npx skills add Peiiii/nextclaw --skill nextclaw-http-agent-runtime-integration -a claude-code`. Or copy the skill folder (.agents/wiki/skills/architecture/nextclaw-http-agent-runtime-integration in Peiiii/nextclaw) into .claude/skills/nextclaw-http-agent-runtime-integration in your project. Claude Code loads it when a task matches its description.

How do I install Nextclaw HTTP Agent Runtime Integration in Codex?

Run `npx skills add Peiiii/nextclaw --skill nextclaw-http-agent-runtime-integration -a codex`. Or copy the skill folder (.agents/wiki/skills/architecture/nextclaw-http-agent-runtime-integration in Peiiii/nextclaw) into .agents/skills/nextclaw-http-agent-runtime-integration in your project. Codex loads it when a task matches its description.

Can I use Nextclaw HTTP Agent Runtime Integration in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Peiiii/nextclaw --skill nextclaw-http-agent-runtime-integration -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/nextclaw-http-agent-runtime-integration, .gemini/skills/nextclaw-http-agent-runtime-integration, .github/skills/nextclaw-http-agent-runtime-integration and .opencode/skills/nextclaw-http-agent-runtime-integration in your project.

What does Nextclaw HTTP Agent Runtime Integration need to run?

Going by SKILL.md and its folder, Nextclaw HTTP Agent Runtime Integration needs the command-line tools its instructions call (pnpm).

Does Nextclaw HTTP Agent Runtime Integration access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Nextclaw HTTP Agent Runtime Integration safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Nextclaw HTTP Agent Runtime Integration use?

Nextclaw HTTP Agent Runtime Integration is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Nextclaw HTTP Agent Runtime Integration use?

About 1.9k tokens (SKILL.md is roughly 7.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Nextclaw HTTP Agent Runtime Integration?

Skills that share tags, products or a category with Nextclaw HTTP Agent Runtime Integration: Trellis Session Insight (mindfold-ai/Trellis, 15k stars), Native Data Fetching (CherryHQ/cherry-studio-app, 4k stars), Debugging Executions (n8n-io/n8n, 207k stars) and Aoti Debug (pytorch/pytorch, 104k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Nextclaw HTTP Agent Runtime Integration?

Peiiii (a GitHub user) maintains it in Peiiii/nextclaw, which has 260 GitHub stars. The repository holds 70 skills in this directory. The repository was last updated on October 7, 2026.

Source: Peiiii/nextclaw on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.