Broker Api Deprecation Notice Monitoring logo

Broker Api Deprecation Notice Monitoring

Community
HimanshuJ16
broker-api-deprecation-notice-monitoring

Use when a long-running bot calls REST endpoints a broker may retire; parses RFC 8594 Sunset and RFC 9745 Deprecation headers plus changelog feeds into one dated migration deadline with an escalation tier.

Overview

PublisherHimanshuJ16
RepositoryAlgo-Trading-Skills
Skill namebroker-api-deprecation-notice-monitoring
Stars
50
Forks
7
Bundled files
5
LicenseApache-2.0
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.

  • 5 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by HimanshuJ16 on GitHub. Read the source before you install it.

Installation

Install the Broker Api Deprecation Notice Monitoring 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/HimanshuJ16/Algo-Trading-Skills/tree/main/skills/broker-api-deprecation-notice-monitoring
  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/HimanshuJ16/Algo-Trading-Skills.git /tmp/Algo-Trading-Skills
mkdir -p .claude/skills
cp -r /tmp/Algo-Trading-Skills/skills/broker-api-deprecation-notice-monitoring .claude/skills/broker-api-deprecation-notice-monitoring
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Broker Api Deprecation Notice Monitoring 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 Broker Api Deprecation Notice Monitoring 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 Broker Api Deprecation Notice Monitoring 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.

When to Use

Invoke this skill when operating long-running algorithmic trading systems against REST/WebSocket broker APIs. Brokers retire legacy endpoints and change payload schemas, and the announcement usually arrives weeks ahead — in a developer changelog, and sometimes as an RFC 8594 Sunset or RFC 9745 Deprecation response header. This skill turns those scattered signals into a dated, de-duplicated migration deadline with an escalation tier attached.

The monitor is a deadline detector, and its failure modes are asymmetric: a false positive costs an engineer a few minutes reading a changelog entry, a false negative means a bot keeps calling an endpoint that stops answering mid-session. Every classification decision below is biased accordingly.

When NOT to Use

  • As proof that nothing is being retired. Header adoption is inconsistent across brokers, and RFC 8594 Section 3 is explicit that Sunset timestamps are hints: "it is not guaranteed that the resource will, in fact, be available until that time and will not be available after that time." An empty registry means nothing was signalled, not that nothing is changing. Alpaca's Market Data API v1 retirement, for instance, was announced in a blog post, not a header.
  • As the migration itself. This detects and dates a deadline. Executing the version cutover belongs to broker-api-versioning-migration-playbook; detecting schema-level breakage belongs to broker-api-changelog-diffing-tool.
  • For live outage detection. A sunset date is a scheduled retirement, not an unplanned outage. Use broker-status-page-monitoring-integration.
  • As an authoritative parse of changelog prose. The changelog path is a keyword-and-date heuristic that routes a human to the entry. Do not auto-disable a strategy on its output alone.

Prerequisites

  • A broker API adapter that exposes HTTP response headers to a hook or middleware layer.
  • The broker's developer changelog RSS/Atom/JSON feed URL, and a scheduler to poll it.
  • A structured logging pipeline or alert router (PagerDuty, Slack, Opsgenie).

Workflow

  1. Inspect live response headers. Hook inspect_http_headers into the HTTP client's global response handler and check every response for Sunset (RFC 8594), Deprecation (RFC 9745), Link with rel="sunset" or rel="deprecation" (RFC 8288), and the non-standard X-API-Deprecation-Warning. This hook runs on the same thread as live orders, so it must never raise into the caller — contain unexpected failures, log them with a traceback, and return no notice.

  2. Parse the two headers by their different contracts. Sunset is an HTTP-date, so accept all three HTTP-date forms (IMF-fixdate, obsolete RFC 850, asctime). RFC 9745 Deprecation is a Structured Field Date — @1688169599, seconds since the epoch — but deployed gateways still emit the pre-RFC Deprecation: true, which asserts deprecation without dating it. Treat an unparseable Deprecation value as "deprecated, date unknown", never as "not deprecated".

  3. Keep deprecation and sunset distinct. Deprecation is when the broker stops recommending the endpoint; sunset is when it stops answering. Derive urgency only from the sunset date. If Sunset is earlier than Deprecation, the broker's own metadata violates RFC 9745 Section 4 — log it and confirm the dates before planning a migration around either.

  4. Poll the developer changelog. Match deprecation keywords, then extract every unambiguous date. Discard candidates that fall before the entry's publication date — those are publication dates and release history, not deadlines — and take the earliest of what survives, because a later date in the same entry is usually the support window for the replacement API. Do not parse locale-ambiguous numeric dates (11/12/2026): day-first and month-first are indistinguishable without knowing the publisher, and guessing wrong moves a deadline by up to eleven months.

  5. Classify urgency from a single clock reading. With $D$ = whole days remaining, floored, computed in UTC:

    • Sunset instant already passed → EXPIRED.
    • $D \le 7$ → CRITICAL_SUNSET_IMMINENT.
    • $7 < D \le 30$ → WARNING_30_DAYS.
    • $D > 30$, or no sunset date at all → NOTICE.

    Expiry is decided by comparing instants, not by testing whether the floored day count reached zero. A sunset 23 hours away has zero whole days left and has not expired; reporting it as EXPIRED tells the desk the migration window is gone while a final day of it remains. The 7/30 day thresholds are operational conventions for migration lead time — neither RFC mandates them — so they are configurable.

  6. De-duplicate, then route. Key the registry by broker, endpoint, and entry identity; alert on a new notice, an escalation, or a broker moving the sunset date, and suppress pure de-escalation so routine re-inspection does not page anyone. Route NOTICE to logs, WARNING_30_DAYS to Slack/Teams, and CRITICAL_SUNSET_IMMINENT/EXPIRED to the on-call rota.

Full procedure: see references/workflows.md. Standards reference: see references/standards.md. Printable pre-flight checklist: see assets/checklist.md.

Common Pitfalls

  • Crashing the request path with the monitor. A header hook runs inline with every live order and quote request. Wiring now_fn=datetime.datetime.now (naive) makes the sunset subtraction raise TypeError on the first deprecated endpoint — converting "this breaks in 30 days" into "this breaks now". Interpret naive clocks as UTC, and contain unexpected errors rather than propagating them.
  • Taking the first date found in a changelog entry. Entries lead with their publication date and often cite the replacement API's support window. Picking the first match dates the deadline to the announcement; picking the latest dates it to the replacement's end of life. Either way a real three-week deadline can be filed as NOTICE or as already-EXPIRED — a silent miss in the one place this skill exists to prevent one.
  • Reading days_remaining == 0 as expired. timedelta.days floors, so a sunset 0.96 days away reports 0. Compare instants for expiry and keep the floored count for display only.
  • Collapsing every changelog notice onto one key. Keying feed notices by broker:GLOBAL_FEED means each new announcement evicts the previous one, so a broker retiring three endpoints shows one outstanding deprecation.
  • Assuming Deprecation: true is still the standard. RFC 9745 (Standards Track, March 2025) requires a Structured Field Date; a parser written only for the boolean form silently drops the date the broker actually supplied.
  • Splitting a Link header on commas. Commas are legal inside a link target, so <https://x/docs?tags=v1,legacy>; rel="sunset" splits into two fragments and the migration URL is lost. Match <target> forms directly, and compare rel as a whole token — RFC 8288 allows several space-separated relation types in one parameter.
  • Timing the countdown twice. Computing days_remaining and the urgency tier from two separate clock readings lets them straddle a threshold and disagree.
  • Monitoring REST only. WebSocket feeds are retired too, and their deprecation notices arrive in control frames or the changelog, never in a REST response header.

Verification

  • Run python -m unittest discover -s skills/broker-api-deprecation-notice-monitoring/scripts and confirm all tests pass.
  • Feed a mock response carrying Sunset: Wed, 25 Nov 2026 00:00:00 GMT against a frozen clock and confirm the extracted UTC date, the day count, and the escalation tier.
  • Confirm the boundary cases explicitly: a sunset 23 hours out is CRITICAL, not EXPIRED; exactly 7 days is CRITICAL; exactly 30 days is WARNING_30_DAYS.
  • Feed a changelog entry whose body restates its publication date before naming a later sunset date, and confirm the later date is chosen.
  • Hand the header inspector None, a list, and {"Sunset": None}, and confirm it returns no notice and raises nothing.

Related Skills

  • broker-api-versioning-migration-playbook
  • broker-api-changelog-diffing-tool
  • broker-status-page-monitoring-integration
  • structured-logging-for-post-incident-forensics

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Broker Api Deprecation Notice Monitoring AI skill do?

Use when a long-running bot calls REST endpoints a broker may retire; parses RFC 8594 Sunset and RFC 9745 Deprecation headers plus changelog feeds into one dated migration deadline with an escalation tier.

Why use Broker Api Deprecation Notice Monitoring on TypingMind?

Because you install it once and use it with any model. Broker Api Deprecation Notice Monitoring 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 Broker Api Deprecation Notice Monitoring in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/HimanshuJ16/Algo-Trading-Skills/tree/main/skills/broker-api-deprecation-notice-monitoring. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Broker Api Deprecation Notice Monitoring?

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 Broker Api Deprecation Notice Monitoring?

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

Is the Broker Api Deprecation Notice Monitoring AI skill free?

Yes. It is published on GitHub by HimanshuJ16 under the Apache-2.0 license. 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 👇