Add mandatory MCP health gate with abort-on-unavailability to all 10 agentic workflows by Copilot · Pull Request #465 · Hack23/riksdagsmonitor

Copilot · 2026-02-23T07:24:32Z

Agentic workflows called get_sync_status() as a warm-up hint but never enforced abort on failure — agents could fall back to analyzing existing repo articles or fabricating content when MCP was unreachable, silently committing recycled/hallucinated output.

Changes

All 10 `news-*.md` workflow definitions

Added ## MANDATORY MCP Health Gate section immediately after ## MANDATORY Date Validation in each file:

## MANDATORY MCP Health Gate

Before generating ANY articles, verify MCP connectivity:

1. Call `get_sync_status({})` — if successful, proceed
2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
3. If ALL 3 attempts fail:
   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
   - DO NOT analyze existing articles in the repository
   - DO NOT fabricate or recycle content
   - The workflow MUST end with noop

**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
- Existing articles in the news/ directory
- Cached or stale data
- AI-generated content without MCP source data

`news-realtime-monitor.md` — contradiction fix

The existing Error Handling section previously said "MCP server unavailable … let workflow FAIL (don't use noop)", directly contradicting the new health gate. Updated to: "Follow MCP Health Gate — retry 3 times, then call safeoutputs___noop".

`sitemap.xml` — merge conflict

Our branch hadn't touched this generated file; resolved by bringing it up to date with main.

Lock files (*.lock.yml) do not need recompilation — they use {{#runtime-import}} to load the .md body dynamically at runtime.

Original prompt

This section details on the original issue you should resolve

<issue_title>Agentic workflows: Add MCP health gate with abort on unavailability</issue_title>
<issue_description>## 📋 Issue Type
Feature / Agentic Workflow Improvement

🎯 Objective

Update all agentic workflow definition files (.github/workflows/news-*.md) to add explicit MCP health check validation with abort behavior. Currently, the workflows call get_sync_status() but don't enforce failure if MCP is unreachable — the agent can still proceed to generate articles from existing file analysis or fabricate content.

📊 Current State

In all 10 workflow .md files:

get_sync_status({}) is called as warm-up but failure is not enforced

The workflow prompt says "ALWAYS call get_sync_status() FIRST" but doesn't say "ABORT if it fails"

Agent can fall back to analyzing existing articles in the repo when MCP is unavailable

This leads to fabricated or recycled content being committed as new articles

No explicit instruction to use noop when MCP is truly unavailable

🚀 Desired State

Each workflow .md file should include:

Explicit MCP health gate: "If get_sync_status() fails after 3 retries, use safeoutputs___noop with message 'MCP server unavailable' and DO NOT generate any articles"

Anti-fabrication rule: "NEVER generate articles by analyzing existing articles in the repo. ALL content MUST come from live MCP data"

Connection retry instructions: "Try get_sync_status() up to 3 times with 30-second waits between attempts before declaring MCP unavailable"

Clear abort path: "If MCP is unavailable: noop. If MCP returns empty data: noop. If MCP returns data: generate articles"

🔧 Implementation Approach

Target files: .github/workflows/news-*.md (all 10 workflow definition files)

Add a new section after "MANDATORY Date Validation" in each workflow:
## MANDATORY MCP Health Gate

Before generating ANY articles, verify MCP connectivity:

1. Call `get_sync_status({})` — if successful, proceed
2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
3. If ALL 3 attempts fail:
   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
   - DO NOT analyze existing articles in the repository
   - DO NOT fabricate or recycle content
   - The workflow MUST end with noop

**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
- Existing articles in the news/ directory
- Cached or stale data
- AI-generated content without MCP source data
✅ Acceptance Criteria

All 10 workflow .md files include MCP health gate section

Clear 3-retry-then-abort instruction in each workflow

Anti-fabrication rule explicitly stated

noop used correctly when MCP unavailable

Lock files recompiled after workflow changes

Consistent wording across all 10 workflows

📚 References

.github/workflows/news-motions.md

.github/workflows/news-propositions.md

.github/workflows/news-committee-reports.md

.github/workflows/news-evening-analysis.md

.github/workflows/news-realtime-monitor.md

.github/workflows/news-week-ahead.md

.github/workflows/news-month-ahead.md

.github/workflows/news-weekly-review.md

.github/workflows/news-monthly-review.md

.github/workflows/news-article-generator.md

🤖 Recommended Agent

Copilot coding agent — update all 10 workflow .md files with MCP health gate</issue_description>

<agent_instructions>Focus ONLY on the .github/workflows/news-*.md files (all 10 workflow definition files). Add a new section called 'MANDATORY MCP Health Gate' after the 'MANDATORY Date Validation' section in EACH file. The section should instruct the agent to: (1) call get_sync_status() up to 3 times with 30-second waits, (2) if all 3 fail, use safeoutputs___noop with message about MCP unavailability, (3) NEVER generate articles from existing repo content or fabricated data. Also add an anti-fabrication rule. Make the wording consistent across all 10 files. Do NOT modify any TypeScript files (scripts/), test files, or lock.yml files to avoid merge conflicts.</agent_instructions>

Comments on the Issue (you are @copilot in this section)

Fixes Agentic workflows: Add MCP health gate with abort on unavailability #464

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

Co-authored-by: pethers <[email protected]>

…agentic workflows Co-authored-by: pethers <[email protected]>

Copilot

Pull request overview

This PR strengthens the repository’s agentic news workflows by adding a mandatory MCP connectivity “health gate” so workflows abort early (via noop) when the MCP server is unavailable, preventing generation from repo content or fabricated sources.

Changes:

Added a standardized “MANDATORY MCP Health Gate” section to all 10 news-*.md workflow definitions.
Introduced explicit 3-attempt retry guidance and a required noop abort path on MCP unavailability.
Added explicit anti-fabrication / anti-recycling instructions tying all article content to live MCP data.

Reviewed changes

Copilot reviewed 10 out of 11 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
.github/workflows/news-weekly-review.md	Adds MCP health gate before generation to enforce MCP-sourced content only.
.github/workflows/news-week-ahead.md	Adds MCP health gate + noop abort path prior to prospective generation.
.github/workflows/news-realtime-monitor.md	Adds MCP health gate to prevent breaking-news generation without live MCP connectivity.
.github/workflows/news-propositions.md	Adds MCP health gate to prevent proposition articles without live MCP data.
.github/workflows/news-motions.md	Adds MCP health gate to prevent motions articles without live MCP data.
.github/workflows/news-monthly-review.md	Adds MCP health gate to prevent monthly review generation without live MCP data.
.github/workflows/news-month-ahead.md	Adds MCP health gate to prevent month-ahead generation without live MCP data.
.github/workflows/news-evening-analysis.md	Adds MCP health gate to prevent evening analysis generation without live MCP data.
.github/workflows/news-committee-reports.md	Adds MCP health gate to prevent committee report articles without live MCP data.
.github/workflows/news-article-generator.md	Adds MCP health gate to prevent manual multi-type generation without live MCP data.

Comments suppressed due to low confidence (9)

.github/workflows/news-week-ahead.md:138

The MCP health gate adds a required noop when MCP is unreachable, but the later PR creation rule says noop is ONLY allowed if there are genuinely no upcoming calendar events. In the MCP-unavailable scenario the agent can’t validate that condition, so the instructions conflict. Adjust the PR creation/noop rules to explicitly permit noop when the MCP health gate fails.

## MANDATORY MCP Health Gate

Before generating ANY articles, verify MCP connectivity:

1. Call `get_sync_status({})` — if successful, proceed
2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
3. If ALL 3 attempts fail:
   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
   - DO NOT analyze existing articles in the repository
   - DO NOT fabricate or recycle content
   - The workflow MUST end with noop

**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
- Existing articles in the news/ directory
- Cached or stale data
- AI-generated content without MCP source data

## MANDATORY PR Creation

- ✅ `safeoutputs___create_pull_request` when articles generated
- ✅ `noop` ONLY if genuinely no upcoming calendar events
- ❌ NEVER use `noop` as fallback for PR creation failures

.github/workflows/news-propositions.md:138

The MCP health gate adds a mandatory noop path when MCP is unreachable, but the later PR creation rule says noop is ONLY allowed if there are genuinely no new propositions. In the MCP-down case the agent can’t confirm that condition, so the rules conflict. Update the PR creation/noop criteria to explicitly allow noop when the MCP health gate fails.

## MANDATORY MCP Health Gate

Before generating ANY articles, verify MCP connectivity:

1. Call `get_sync_status({})` — if successful, proceed
2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
3. If ALL 3 attempts fail:
   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
   - DO NOT analyze existing articles in the repository
   - DO NOT fabricate or recycle content
   - The workflow MUST end with noop

**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
- Existing articles in the news/ directory
- Cached or stale data
- AI-generated content without MCP source data

## MANDATORY PR Creation

- ✅ `safeoutputs___create_pull_request` when articles generated
- ✅ `noop` ONLY if genuinely no new propositions
- ❌ NEVER use `noop` as fallback for PR creation failures

.github/workflows/news-motions.md:138

The MCP health gate mandates noop when MCP is unavailable, but the later PR creation rule says noop is ONLY allowed if there are genuinely no new motions. In the MCP-down case the agent can’t validate that, so the instructions conflict. Update the PR creation/noop rules to explicitly permit noop when the MCP health gate fails.

## MANDATORY MCP Health Gate

Before generating ANY articles, verify MCP connectivity:

1. Call `get_sync_status({})` — if successful, proceed
2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
3. If ALL 3 attempts fail:
   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
   - DO NOT analyze existing articles in the repository
   - DO NOT fabricate or recycle content
   - The workflow MUST end with noop

**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
- Existing articles in the news/ directory
- Cached or stale data
- AI-generated content without MCP source data

## MANDATORY PR Creation

- ✅ `safeoutputs___create_pull_request` when articles generated
- ✅ `noop` ONLY if genuinely no new motions
- ❌ NEVER use `noop` as fallback for PR creation failures

.github/workflows/news-monthly-review.md:140

The MCP health gate introduces a mandatory noop when MCP is unreachable, but the later PR creation rule says noop is ONLY allowed if there is genuinely no parliamentary activity in the past month. In the MCP-down case the agent can’t verify that condition, so the instructions conflict. Update the PR creation/noop criteria to explicitly allow noop when the MCP health gate fails.

## MANDATORY MCP Health Gate

Before generating ANY articles, verify MCP connectivity:

1. Call `get_sync_status({})` — if successful, proceed
2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
3. If ALL 3 attempts fail:
   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
   - DO NOT analyze existing articles in the repository
   - DO NOT fabricate or recycle content
   - The workflow MUST end with noop

**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
- Existing articles in the news/ directory
- Cached or stale data
- AI-generated content without MCP source data

## MANDATORY PR Creation

- ✅ `safeoutputs___create_pull_request` when articles generated
- ✅ `noop` ONLY if genuinely no parliamentary activity in past month
- ❌ NEVER use `noop` as fallback for PR creation failures

.github/workflows/news-month-ahead.md:140

The MCP health gate mandates noop when MCP is unavailable, but the later PR creation rule says noop is ONLY allowed if there are genuinely no upcoming events in the next 30 days. In the MCP-down case the agent can’t validate that condition, so the instructions conflict. Update the PR creation/noop rules to explicitly allow noop when the MCP health gate fails.

## MANDATORY MCP Health Gate

Before generating ANY articles, verify MCP connectivity:

1. Call `get_sync_status({})` — if successful, proceed
2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
3. If ALL 3 attempts fail:
   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
   - DO NOT analyze existing articles in the repository
   - DO NOT fabricate or recycle content
   - The workflow MUST end with noop

**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
- Existing articles in the news/ directory
- Cached or stale data
- AI-generated content without MCP source data

## MANDATORY PR Creation

- ✅ `safeoutputs___create_pull_request` when articles generated
- ✅ `noop` ONLY if genuinely no upcoming events in next 30 days
- ❌ NEVER use `noop` as fallback for PR creation failures

.github/workflows/news-committee-reports.md:145

The MCP health gate mandates noop when MCP is unreachable, but the later PR creation section says noop is ONLY allowed if there are genuinely no committee reports from MCP. With MCP unavailable, the agent can’t confirm that condition, so the instructions conflict. Update the PR creation/noop rules to explicitly allow noop when the MCP health gate fails.

## MANDATORY MCP Health Gate

Before generating ANY articles, verify MCP connectivity:

1. Call `get_sync_status({})` — if successful, proceed
2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
3. If ALL 3 attempts fail:
   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
   - DO NOT analyze existing articles in the repository
   - DO NOT fabricate or recycle content
   - The workflow MUST end with noop

**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
- Existing articles in the news/ directory
- Cached or stale data
- AI-generated content without MCP source data

## MANDATORY PR Creation

- ✅ **REQUIRED:** `safeoutputs___create_pull_request` when articles generated
- ✅ **ONLY USE `noop` if genuinely no new committee reports** from riksdag-regering-mcp
- ❌ **NEVER use `noop` as fallback for PR creation failures**

.github/workflows/news-evening-analysis.md:142

This MCP health gate mandates ending with noop when MCP is unavailable, but the subsequent "MANDATORY Pull Request Creation" section says the workflow MUST create a PR with analysis or FAIL and that noop is only for "insufficient data" after checking MCP. With MCP unreachable, the workflow can’t satisfy those later criteria, so the instructions are contradictory. Update the PR creation rules to explicitly allow noop when the MCP health gate fails, while keeping the "PR must succeed if articles were generated" requirement.

## MANDATORY MCP Health Gate

Before generating ANY articles, verify MCP connectivity:

1. Call `get_sync_status({})` — if successful, proceed
2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
3. If ALL 3 attempts fail:
   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
   - DO NOT analyze existing articles in the repository
   - DO NOT fabricate or recycle content
   - The workflow MUST end with noop

**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
- Existing articles in the news/ directory
- Cached or stale data
- AI-generated content without MCP source data

### 2. MANDATORY Pull Request Creation (Final Step)

**CRITICAL: Workflow MUST create a PR with analysis or FAIL**

From a reader's perspective: **Where's the analysis?**

- ✅ **REQUIRED:** `safeoutputs___create_pull_request` - When analysis articles generated
- ✅ **ONLY USE `noop` if genuinely insufficient data** (checked riksdag-regering-mcp, found no votes, no debates, no documents, no calendar events for the lookback period)
- ❌ **NEVER use `noop` as a fallback for PR creation failures** - If articles were generated but PR fails, the workflow MUST FAIL

**⚠️ If you generated analysis articles but cannot create PR → workflow FAILS (not noop)**

.github/workflows/news-weekly-review.md:140

The MCP health gate introduces a new noop path when MCP is unreachable, but the later "MANDATORY PR Creation" section says noop is ONLY allowed when there is genuinely no weekly activity. This creates conflicting instructions for the agent in the MCP-down case (it can’t verify activity but is told to noop). Update the PR creation/noop criteria to explicitly allow noop when the MCP health gate fails (in addition to the normal no-activity case) so the abort behavior is unambiguous.

## MANDATORY MCP Health Gate

Before generating ANY articles, verify MCP connectivity:

1. Call `get_sync_status({})` — if successful, proceed
2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
3. If ALL 3 attempts fail:
   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
   - DO NOT analyze existing articles in the repository
   - DO NOT fabricate or recycle content
   - The workflow MUST end with noop

**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
- Existing articles in the news/ directory
- Cached or stale data
- AI-generated content without MCP source data

## MANDATORY PR Creation

- ✅ `safeoutputs___create_pull_request` when articles generated
- ✅ `noop` ONLY if genuinely no parliamentary activity in the past week
- ❌ NEVER use `noop` as fallback for PR creation failures

.github/workflows/news-article-generator.md:150

This MCP health gate mandates ending with noop when MCP is unavailable, but the following PR creation rules say the workflow MUST create a PR with articles or FAIL and that noop is only for "no new data" after checking MCP. With MCP unreachable, those later criteria can’t be met, so the instructions conflict. Update the PR creation/noop rules to explicitly allow noop when the MCP health gate fails, while still failing if articles were generated but PR creation fails.

## MANDATORY MCP Health Gate

Before generating ANY articles, verify MCP connectivity:

1. Call `get_sync_status({})` — if successful, proceed
2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
3. If ALL 3 attempts fail:
   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
   - DO NOT analyze existing articles in the repository
   - DO NOT fabricate or recycle content
   - The workflow MUST end with noop

**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
- Existing articles in the news/ directory
- Cached or stale data
- AI-generated content without MCP source data

### 2. MANDATORY Pull Request Creation (Final Step)

**CRITICAL: Workflow MUST create a PR with articles or FAIL**

From a reader's perspective: **Where's the article?**

- ✅ **REQUIRED:** `safeoutputs___create_pull_request` - When articles generated
- ✅ **ONLY USE `noop` if genuinely no new data** (checked riksdag-regering-mcp, found no committee reports, no propositions, no significant updates, AND force_generation=false)
- ❌ **NEVER use `noop` as a fallback for PR creation failures** - If articles were generated but PR fails, the workflow MUST FAIL

**⚠️ If you generated articles but cannot create PR → workflow FAILS (not noop)**

Copilot · 2026-02-23T10:51:34Z

.github/workflows/news-realtime-monitor.md

+## MANDATORY MCP Health Gate
+
+Before generating ANY articles, verify MCP connectivity:
+
+1. Call `get_sync_status({})` — if successful, proceed
+2. If it fails, wait 30 seconds and retry (up to 3 total attempts)
+3. If ALL 3 attempts fail:
+   - Use `safeoutputs___noop` with message: "MCP server unavailable after 3 connection attempts. No articles generated."
+   - DO NOT analyze existing articles in the repository
+   - DO NOT fabricate or recycle content
+   - The workflow MUST end with noop
+
+**CRITICAL**: ALL article content MUST originate from live MCP data. Never generate content from:
+- Existing articles in the news/ directory
+- Cached or stale data
+- AI-generated content without MCP source data
+


The MCP health gate mandates ending with safeoutputs___noop if MCP is unavailable, but later in this workflow the Error Handling section states "MCP server unavailable … let workflow FAIL (don't use noop)". These instructions contradict each other; align them by updating/removing the Error Handling guidance so MCP-unavailable consistently results in noop per the health gate.

Fixed in 03f5203. The Error Handling section now reads: "Follow MCP Health Gate — retry 3 times, then call safeoutputs___noop with message 'MCP server unavailable after 3 connection attempts. No articles generated.' (do NOT let the workflow fail)" — consistent with the health gate at the top of the file.

pethers · 2026-02-23T11:50:10Z

@copilot apply changes based on the comments in this thread

…ith MCP Health Gate (noop on unavailability) Co-authored-by: pethers <[email protected]>

Copilot

Pull request overview

Copilot reviewed 10 out of 11 changed files in this pull request and generated no new comments.

Co-authored-by: pethers <[email protected]>

github-actions · 2026-02-23T14:45:41Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

Copilot

Pull request overview

Copilot reviewed 10 out of 10 changed files in this pull request and generated no new comments.

Initial plan

060f247

Copilot AI assigned Copilot and pethers Feb 23, 2026

Copilot started work on behalf of pethers February 23, 2026 07:39 View session

Add MANDATORY MCP Health Gate to all 10 agentic workflow .md files

fdf1b58

Co-authored-by: pethers <[email protected]>

Copilot AI changed the title ~~[WIP] Add MCP health gate with abort on unavailability~~ Add MANDATORY MCP Health Gate with abort-on-unavailability to all 10 agentic workflows Feb 23, 2026

Copilot AI requested a review from pethers February 23, 2026 07:48

Copilot finished work on behalf of pethers February 23, 2026 07:48

Copilot started work on behalf of pethers February 23, 2026 09:47 View session

Add mandatory MCP health gate with abort-on-unavailability to all 10 …

d1a17ab

…agentic workflows Co-authored-by: pethers <[email protected]>

Copilot AI changed the title ~~Add MANDATORY MCP Health Gate with abort-on-unavailability to all 10 agentic workflows~~ Add mandatory MCP health gate with abort-on-unavailability to all 10 agentic workflows Feb 23, 2026

Copilot finished work on behalf of pethers February 23, 2026 09:56

pethers requested a review from Copilot February 23, 2026 10:47

Copilot started reviewing on behalf of pethers February 23, 2026 10:48 View session

Copilot AI reviewed Feb 23, 2026

View reviewed changes

Copilot started work on behalf of pethers February 23, 2026 11:50 View session

Fix contradiction in news-realtime-monitor.md: align Error Handling w…

03f5203

…ith MCP Health Gate (noop on unavailability) Co-authored-by: pethers <[email protected]>

Copilot finished work on behalf of pethers February 23, 2026 11:55

pethers requested a review from Copilot February 23, 2026 12:37

Copilot started reviewing on behalf of pethers February 23, 2026 12:37 View session

Copilot AI reviewed Feb 23, 2026

View reviewed changes

Copilot started work on behalf of pethers February 23, 2026 13:32 View session

Resolve merge conflict: update sitemap.xml to match main

c9caabc

Co-authored-by: pethers <[email protected]>

Copilot finished work on behalf of pethers February 23, 2026 13:37

Merge branch 'main' into copilot/add-mcp-health-gate

052fc64

pethers requested a review from Copilot February 23, 2026 14:44

github-actions bot added the size-m Medium change (50-250 lines) label Feb 23, 2026

Copilot started reviewing on behalf of pethers February 23, 2026 14:45 View session

Copilot AI reviewed Feb 23, 2026

View reviewed changes

pethers marked this pull request as ready for review February 23, 2026 15:59

pethers merged commit 33446a8 into main Feb 23, 2026
15 checks passed

pethers deleted the copilot/add-mcp-health-gate branch February 23, 2026 16:00

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Comments

Add mandatory MCP health gate with abort-on-unavailability to all 10 agentic workflows#465

Add mandatory MCP health gate with abort-on-unavailability to all 10 agentic workflows#465
pethers merged 6 commits intomainfrom
copilot/add-mcp-health-gate

Copilot AI commented Feb 23, 2026 •

edited

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Feb 23, 2026

Uh oh!

Copilot AI Feb 23, 2026

Uh oh!

pethers commented Feb 23, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

github-actions bot commented Feb 23, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Comments

Conversation

Copilot AI commented Feb 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changes

All 10 news-*.md workflow definitions

news-realtime-monitor.md — contradiction fix

sitemap.xml — merge conflict

🎯 Objective

📊 Current State

🚀 Desired State

🔧 Implementation Approach

✅ Acceptance Criteria

📚 References

🤖 Recommended Agent

Comments on the Issue (you are @copilot in this section)

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Feb 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 23, 2026

Choose a reason for hiding this comment

Uh oh!

pethers commented Feb 23, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

github-actions bot commented Feb 23, 2026

🔍 Lighthouse Performance Audit

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Copilot AI commented Feb 23, 2026 •

edited

Loading

All 10 `news-*.md` workflow definitions

`news-realtime-monitor.md` — contradiction fix

`sitemap.xml` — merge conflict