Wiring External Webhooks logo

Wiring External Webhooks

Organization
friday-platform
wiring-external-webhooks

Connects an external service's webhook (Bitbucket / Jira / GitHub / custom) to a workspace's HTTP signal via Friday's `/hook/raw/` tunnel URL. Use ONLY when the user asks how to point an upstream system at Friday: the URL shape, how to get the current tunnel host, the upstream UI steps, or the workspace.yml signal shape that accepts the body without stripping it. Everything that happens AFTER Friday receives the webhook — agent implementation, payload parsing, env wiring, HMAC verification, posting back to the upstream — is out of scope for this skill.

Overview

Publisherfriday-platform
Repositoryfriday-studio
Skill namewiring-external-webhooks
Stars
102
Forks
5
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by friday-platform on GitHub. Read the source before you install it.

Installation

Install the Wiring External Webhooks AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/friday-platform/friday-studio/tree/main/packages/system/skills/wiring-external-webhooks
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/friday-platform/friday-studio.git /tmp/friday-studio
mkdir -p .claude/skills
cp -r /tmp/friday-studio/packages/system/skills/wiring-external-webhooks .claude/skills/wiring-external-webhooks
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Wiring External Webhooks in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Wiring External Webhooks on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Wiring External Webhooks is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Wiring external webhooks to HTTP signals

A provider: http signal in workspace.yml exposes a URL. To make something fire it you register the URL with the upstream system (any service that can POST JSON). Friday's webhook-tunnel is the public entry point — it forwards the request body to atlasd's signal endpoint.

The tunnel does one thing: take the POST body and hand it to the workspace's signal. No HMAC verification, no event filtering, no field extraction. Every webhook from every upstream uses the same URL shape and the same body-passthrough semantics.

URL pattern — always /hook/raw/ on the PUBLIC tunnel host

https://<public-tunnel-url>/hook/raw/<workspaceId>/<signalId>
  • <public-tunnel-url> is the cloudflared trycloudflare URL — fetch it live from /status (next section). It rotates on tunnel restart.
  • <workspaceId> is the runtime id (e.g. light_papaya, not the friendly name).
  • <signalId> is the signal's key in workspace.yml.

Never give an external service a URL of the form <friday-host>/api/workspaces/<wsId>/signals/<signalId> — that path is atlasd's internal direct route. It is not exposed through cloudflared, the upstream cannot reach it from the public internet, and recommending it is the original foot-gun that motivated this skill. If you cannot reach /status from your environment to fetch the real tunnel URL (run_code does not always have network access to the local tunnel), STOP and ask the user to run the /status curl themselves and paste the URL back — do not synthesize a host or fall back to the /api/workspaces/... path.

Get the tunnel URL in one call

The local webhook-tunnel listens on whatever $EXTERNAL_TUNNEL_URL points at (the launcher / playground exports this; default is https://localhost:9090 on bare-dev rigs). Always read it from the env — do not hardcode the port, it differs per rig:

bash
curl -sk "${EXTERNAL_TUNNEL_URL:-https://localhost:9090}/status" | jq
# → { "url": "https://...trycloudflare.com",
#     "providers": ["raw"], "pattern": "/hook/raw/{workspaceId}/{signalId}", ... }

If that returns empty (exit 0 but no body), the port is wrong for this rig. Run echo "$EXTERNAL_TUNNEL_URL" to confirm what the launcher set, or ps auxww | grep cloudflared to read the --url flag the tunnel was started with — that's the actual local port. Do NOT silently move on; the URL is the whole point of this skill.

/status is the single source of truth — it shows the current trycloudflare URL (rotates on tunnel restart) and confirms the tunnel is up. Run it FIRST when something's wrong before guessing.

Signal schema for raw webhooks — additionalProperties: true or you lose the body

The signal's JSON Schema is enforced by the runtime through Zod, which STRIPS any fields not declared in properties. If you declare nested objects without additionalProperties: true, every nested field arrives as an empty {} and the agent sees nothing useful.

Wrong (what looks documentative but silently destroys the payload):

yaml
signals:
  bb-pr-comment:
    provider: http
    config: { path: /bb-pr-comment }
    schema:
      type: object
      properties:
        actor:        { type: object, description: "Commenter" }
        comment:      { type: object, description: "Comment object" }
        pullrequest:  { type: object, description: "PR object" }
        repository:   { type: object, description: "Repo object" }

The agent's inputs.actor, inputs.comment, etc. all become {} — LLM templates render as empty, Python agents see empty dicts.

Right — let the body through:

yaml
signals:
  bb-pr-comment:
    provider: http
    config: { path: /bb-pr-comment }
    schema:
      type: object
      additionalProperties: true

Or, if you want to document the expected shape without stripping:

yaml
schema:
  type: object
  additionalProperties: true
  properties:
    actor:        { type: object, additionalProperties: true }
    comment:      { type: object, additionalProperties: true }
    pullrequest:  { type: object, additionalProperties: true }
    repository:   { type: object, additionalProperties: true }

Rule of thumb for HTTP webhook signals: every type: object you write must be paired with additionalProperties: true, or it strips silently. The validator does not warn — the agent just sees {}.

Where the payload lands inside the workspace

Once the tunnel forwards the body to atlasd, the workspace's signal fires and the body is available to the triggered agent under ctx.input.config (or ctx.input.raw as a fallback for signal-direct invocations):

python
payload = ctx.input.config or ctx.input.raw

What the agent then does with the payload — extracting fields, branching on event family, posting something back to the upstream — is agent-implementation territory and out of scope for this skill.

Bitbucket Cloud — UI steps

  1. Repo settings → Webhooks → Add webhook
  2. Title: anything (e.g. Friday)
  3. URL: paste https://<tunnel-url>/hook/raw/<workspaceId>/<signalId>
  4. Secret: leave blank — the tunnel does no verification.
  5. Status: Active
  6. Triggers: pick the events you want.

GitHub — UI steps

  1. Repo settings → Webhooks → Add webhook
  2. Payload URL: https://<tunnel-url>/hook/raw/<workspaceId>/<signalId>
  3. Content type: application/json (not the form-encoded default).
  4. Secret: leave blank — the tunnel does no verification.
  5. Which events: pick the events you want.

Tunnel-level errors

Tunnel responseCauseFix
400 Unknown provider: X. Available: rawWebhook URL uses anything other than /hook/raw/....Change the upstream's URL to /hook/raw/<wsId>/<signalId>.
400 raw provider expects JSON object, got ...Upstream is posting non-JSON or a JSON array.Configure the upstream to send application/json with a top-level object body.
400 empty bodyUpstream is posting with no body.Confirm the upstream is actually sending the event payload.
413 body exceeds N bytesWebhook body bigger than the tunnel's max.Reduce the body upstream.
502 Cannot reach atlasdatlasd unreachable from the tunnel.Check atlasd /health and the JetStream stream.

Frequently asked questions

What does the Wiring External Webhooks AI skill do?

Connects an external service's webhook (Bitbucket / Jira / GitHub / custom) to a workspace's HTTP signal via Friday's `/hook/raw/` tunnel URL. Use ONLY when the user asks how to point an upstream system at Friday: the URL shape, how to get the current tunnel host, the upstream UI steps, or the workspace.yml signal shape that accepts the body without stripping it. Everything that happens AFTER Friday receives the webhook — agent implementation, payload parsing, env wiring, HMAC verification, posting back to the upstream — is out of scope for this skill.

Why use Wiring External Webhooks on TypingMind?

Because you install it once and use it with any model. Wiring External Webhooks is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Wiring External Webhooks in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/friday-platform/friday-studio/tree/main/packages/system/skills/wiring-external-webhooks. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Wiring External Webhooks?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Wiring External Webhooks?

As many as you like. As long as a model supports skills, you can use Wiring External Webhooks with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Wiring External Webhooks AI skill free?

It is published on GitHub by friday-platform. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇