🚀 Agentics Beyond Code

September 17, 2026 · View on GitHub

Agentic Workflows for PMs, ops, compliance, and other non-engineering roles — built on GitHub Agentic Workflows, now in public preview.

While The Agentics focuses on engineering use cases (CI, code review, testing), Agentics Beyond Code brings the same power to the people who ship, govern, and operate products — without writing a line of code.

⏸️ Scheduled workflows are currently paused to reduce API costs while this repo is in demo/reference mode. Workflows triggered by human activity (issue creation, Slack reactions, transcript pushes) remain active. To run the full system, trigger the Sample Data Simulator and Sample Data Launch Creator manually first, then follow the stage run order. To re-enable scheduled runs, uncomment the schedule: lines in each workflow's .md file and recompile with gh aw compile.

🎯 Who is this for?

DRIs / Product Managers — track launches, monitor feature health, keep roadmaps honest
Downstream / Compliance Teams — domain sign-offs, audit trails, policy checks
Leaders — launch pipeline visibility, risk dashboards, trend analysis

📂 Available Workflows

🚢 Launch Tracking

Workflow	Description	Example output
🚦 Launch Readiness Checker	Monday morning readiness report across all launches — completeness, risk, blockers, sign-offs	Readiness report - 2026-05-31

✅ Compliance

Workflow	Description	Example output
🛡️ Compliance Review	Evaluates launches against Security, Privacy, Accessibility, and Responsible AI rubrics — updates labels, posts status tables, creates review sub-issues (Monday mornings)	Status table comment - 2026-05-31
📊 Compliance Team Reports	Monday morning per-team discussion showing launches needing review, sorted by urgency	Security, Privacy, Accessibility, and Responsible AI - week of 2026-06-01

📣 Go-to-Market

Workflow	Description	Example output
📣 GTM Content	Monday morning generation and refresh of changelog announcement drafts and public roadmap items as sub-issues, following the org's voice & tone policy	Changelog draft - EU Payment Methods
📣 GTM Team Reports	Monday morning report summarizing launches needing GTM action — missing changelog drafts, missing roadmap items, content needing refresh, and upcoming launches	GTM readiness report - 2026-05-31

📥 Intake & Triage

Two of the workflows below read Slack, but from different surfaces: Slack Reaction Intake reads the team's own internal workspace (teammates asking the team for something); Customer Feedback Intake reads separate field/CS-facing channels relaying external customer signal, alongside Discord and OSS-repo fixtures. See Two Slack Surfaces for the full comparison.

Workflow	Description	Example output
📥 Intake Request Triage	Scores incoming feature requests and bug reports using RICE and Kano frameworks, checks strategy alignment, detects duplicates, flags incomplete submissions, and adds items to the triage project board	Triage comment
🗣️ Customer Feedback Intake	Manual deterministic fixture intake for external, customer-facing signal — normalizes OSS repo, Discord, and field/CS Slack channel fixtures, preserves exact customer language, creates feedback intake issues, and adds them to the Customer Feedback Queue project	Feedback intake issue #311
🔎 Feedback Dedupe & Strategy Triage	Reviews feedback intake issues, proposes duplicate clusters (linked as GitHub sub-issues of a canonical issue), compares exact customer language against strategy, suggests priority, comments on issues, labels the queue, and updates Customer Feedback Queue fields	Triage comment - #311
📈 Friday Feedback Trends Report	Friday morning discussion recommending which feedback-driven work to consider accepting next week — ranked by suggested priority, confidence, and strategy fit, cross-referenced against active launches/initiatives so it never recommends duplicate work	Feedback Trends - week of 2026-09-14
🏗️ Feedback Work Item Converter	Comment `/create-work-item` on a PM-approved feedback issue to convert it into a clean, agent-ready work item in the Launch Tracker project — redacts customer identity while preserving exact language, links back to every issue in the duplicate cluster	Work item #328, converted from feedback #319

📋 Decision & Knowledge

Workflow	Description	Example output
🔍 Assumption Surfacer	Scans issues for implicit assumptions (timelines, dependencies, user behavior, capacity) and posts them as explicit questions for the team to reason through together	Assumptions comment
😤 Adversarial PM	Wednesday morning grumpy challenge of the week's most consequential decisions — picks 2-3 from `/decisions/`, argues against them using non-deterministic lenses (pre-mortem, reversibility, opportunity cost, etc.), and posts sarcastic but specific counterarguments on the source issues	ClickHouse challenge
📋 Decision Log	Daily scan of issue comments and meeting transcripts for decisions — creates a PR with individual decision record files in `/decisions/`	Decision log - 2026-05-09
🧭 Strategy Alignment	Wednesday morning analysis of team activity against `docs/strategy.md` tradeoffs — comments on clearly misaligned issues, annotates the strategy doc with alignment evidence and emerging patterns	Strategy evidence PR
🎙️ Transcript Processor	Triggered when `.txt` or `.vtt` files are pushed to `/transcripts/` — matches transcript content to open issues and posts summary comments	Meeting notes comment
💬 Slack Context Processor	Pulls relevant Slack conversations into GitHub — matches messages to open issues and posts summary comments with source context and permalinks, so nothing discussed in Slack gets lost	Slack update comment; Slack message (join Slack)
📥 Slack Reaction Intake	React with `:inbox_tray:` on any message in the team's own internal Slack workspace to instantly capture it as a labeled GitHub intake issue — a teammate asking the team for something, not customer feedback — no copy-paste, no context switching, no requests dropped	Slack intake issue; Slack message (join Slack)
🔔 Slack Triage Postback	Closes the loop with Slack reporters — after triage completes, posts the outcome (RICE/Kano scores, alignment verdict, or needs-more-info) directly back into the originating Slack thread	Triaged issue with postback sent
📢 Slack Report-Back	After any reporting workflow completes (Weekly Status, Launch Readiness, Workflow Health, etc.), posts a short link to the generated artifact in the configured Slack channel — keeping Slack-native teammates in the loop without requiring them to watch GitHub	Slack message (join Slack); Weekly Status discussion

📊 Leadership

Workflow	Description	Example output
📋 Weekly Status	Friday morning leadership status rollup — What Shipped, What We Learned, FYI, and SOS — across all initiatives and launches. Automatically generates a collaborative Google Doc draft from the Discussion, posts a finalization gate comment on the Doc, and notifies the team in Slack. Resolving the gate comment triggers staged publishing.	Weekly status - week of 2026-06-01
📋 Leadership Briefs	Monday morning personalized briefs — one per leader policy file — with Give Kudos, Give Feedback, and Get Involved sections tailored to each leader's domain, goals, and management style	Alex Chen and Priya Sharma - week of 2026-06-01

📅 Calendar Intelligence

Fixture-first: all three workflows run against google-calendar-fixtures/ without any credentials configured. Connect a shared team Google Calendar to go live — see google-calendar-fixtures/README.md.

Workflow	Description	Example output
📅 Calendar Load Report	Friday fragmentation score and deep work block analysis per contributor — flags high meeting dispersion and days with zero 90-minute uninterrupted blocks	Calendar Load — week of 2026-06-09
🧭 Calendar Strategy Audit	Wednesday audit classifying team calendar time against `docs/strategy.md` priorities — surfaces which stated priorities have zero calendar coverage and whether async GitHub activity fills those gaps	Calendar Audit — week of 2026-06-09
🔁 Daily Standup Prep	Monday/Wednesday standup prep — posts a discussion with high-priority topics, blockers, and facilitation prompts, enriched with Mode A/B meeting briefs from today's calendar events	Standup prep - 2026-05-31

🔧 Operations

Workflow	Description	Example output
🩺 Workflow Health	Friday morning health report across all agentic workflows — success rates, failure patterns, cost estimates, cross-workflow interaction analysis (conflict detection, cascade chains, resource contention), and efficiency recommendations	Health report - 2026-05-31
🧾 Commitment Reconciler	Monday commitment audit that compares transcripts and issue comments against GitHub artifacts, surfacing promised-but-untracked work, stale commitments, artifact drift, and completion mismatches	Commitment reconciliation - week of 2026-06-01
🔄 Process Analyzer	Weekly retro + process analysis — posts a team retrospective discussion, detects process drift in transcripts vs `docs/how-we-work.md`, identifies automation opportunities and gaps, and creates update PRs	Retro - week of 2026-05-25
🐒 Chaos Monkey	On-demand organizational chaos injection — scores the team's stasis across 6 signals (decision diversity, participation entropy, process staleness, topic homogeneity, launch concentration), and when things are too comfortable, posts a discussion with 2–3 calibrated disruption prescriptions. Stays silent when the team is healthy.	Stasis Report — 2026-06-08

🧪 Demo / Sample Data

Note: The sample data workflows are for demo purposes only. They generate fake project activity so the other workflows have realistic data to work with. You don't need them for production use. The schedules are currently paused — trigger them manually when you want to generate fresh data.

Running workflows manually

When triggering workflows by hand, run them in stages — parallel within each stage, but wait for each stage to complete before starting the next:

Stage	Workflows	Why
1	`sample-data-simulator`, `sample-data-launch-creator`, `customer-feedback-intake`	Generates fresh project and feedback data — must run first
2	`decision-log`, `daily-standup-prep`, `assumption-surfacer`, `process-analyzer`, `compliance-team-reports`, `feedback-dedupe-triage`	Analyze current data
3	`weekly-status`, `leadership-brief`, `friday-feedback-trends-report`	Roll up stage 2 outputs
4	`workflow-health`	Monitor everything — run last

create-work-item isn't part of this staged run — it's triggered on demand by commenting /create-work-item on a PM-approved feedback issue, not run in bulk.

Workflow	Cadence	Description	Example output
🎲 Sample Data Simulator	Daily	Closes tasks, adds progress comments, generates standup transcripts, and creates intake issues — keeps daily activity flowing.	Closed #69 and #176, progress comments on #96 and #262, standup transcript PR #286, intake issue #287
🏗️ Sample Data Launch Creator	Weekly	Creates new launches with epics and tasks, advances launch phases, closes completed launches, and adjusts risk levels — grows the project hierarchy over time.	Created [Launch] Real-Time Notifications #280 with epics #281 and #282
🗣️ Customer Feedback Intake	Manual	Creates Customer Feedback Queue issues from fixture data so feedback triage and Friday reports have realistic cross-channel signals.	Feedback intake issue #311

💡 Philosophy

Six ideas shaped this project:

Process as Code

Team policies — how we triage, what "launch-ready" means, compliance rubrics — are written as markdown files that humans review and agents execute. When the policy file is the automation input, docs can't rot: changing how-we-work.md changes how the workflow behaves on its next run.

Living Documents

Strategy docs, decision logs, and how-we-work guides go stale the moment someone merges a PR or wraps up a meeting. Agentic workflows close that gap by connecting documents to the events that should update them — transcript pushes, issue closures, weekly cadences — so the document stays as current as the work itself. The update always arrives as something reviewable — a PR, a comment, a draft discussion — so the team has a natural moment to push back before anything is final.

Artifacts over Roles

The scoping unit for each workflow is the artifact it produces (a readiness report, a compliance review, a decision record), not the role it replaces. When everyone agrees on the output artifact, the workflow has a natural boundary and the agent's constraints stay focused. Animate the artifact, not the job title. And because the artifact is always something a human can read and act on — not a silent action taken on your behalf — the agent drafts; the team decides.

The Repo Is the Architecture

The model is interchangeable — what makes these workflows effective is the environment they operate in: issues, labels, docs, transcripts, git history. Designing the repo topology well matters more than picking the right LLM.

Your Habits Are Already Triggers

You don't need a new tool to work with agents — the habits you already have are enough. An emoji reaction captures a Slack message as a tracked request. A comment in a Google Doc kicks off a review cycle. A reaction in a thread closes the loop with the person who raised it. The integrations are designed to meet your workflow where it lives, not pull you into another system. When the trigger is something you were already going to do, adoption disappears.

Every Run Improves the System

Good agentic systems compound. The process analyzer detects drift, the decision log accumulates records, strategy alignment annotates docs with evidence. Each run leaves the repo a little smarter than it found it.

🧰 Agent Skills

This repo includes reusable agent skills for setting up and maintaining Agentics Beyond Code workflows:

Non-Coder Agentic Workflow Builder — helps product, ops, compliance, GTM, design, research, support, customer success, program, and leadership users turn process problems into a repo setup with workflows, project boards, issue templates, labels, blank strategy/how-we-work docs, policies, and folders.
Agentic Workflows — helps create, update, debug, compile, and validate GitHub Agentic Workflows. Bundled from the gh-aw framework.
Org Work Sensing — assesses current-state work signals (GitHub, Jira, Linear, docs) and readiness gaps before recommending an agentic workflow setup.
Claude-Native Workflow Builder — the non-GitHub sibling: ports this repo's existing workflows (one, a few, or a whole pipeline), or designs a new one from scratch, as Claude Routines and Scheduled Tasks instead of GitHub Agentic Workflows, for teams whose system of record isn't GitHub or who just want a slice of this repo without the rest of the scaffolding.

Skills have one canonical copy and several discovery views:

.github/skills/ contains the canonical skills and supports GitHub/Copilot discovery.
.agents/skills/ links to the canonical skills for Codex and open Agent Skills-compatible tools.
.claude/skills/ links to the canonical skills for Claude-style discovery.
AGENTS.md provides portable repository guidance for coding agents.

The agentic-workflows skill is maintained upstream by GitHub Next. See docs/skills.md for ownership, refresh, and validation guidance.

📖 Documentation

Getting Started

Getting Started — prerequisites, installation, and first run
How It Works — architecture, issue hierarchy, and customization
FAQ — common questions about setup, workflows, and costs
Workflow Ideas — catalog of future workflow ideas for PM, ops, compliance, and GTM
External Integration Patterns — future work for integrating with Slack, Jira, Microsoft 365, Google Workspace, Salesforce, ServiceNow, Notion, Asana, and Linear
Slack Integration Plan — Slack context ingestion, emoji-driven automation, and Slack report-backs for the team's own internal workspace; three of five phases are built and live, the rest still proposal
Google Docs Integration Plan — fixture-first proposal for bounded Google Docs context reads and validated document updates; the Weekly Status collaborative Google Doc flow (draft → shape → finalize → Slack) is built and was proven live on 2026-06-08 (writeup), currently paused pending an expired Google OAuth token refresh — the broader plan beyond that flow is still proposal

Sample Team Context

These docs represent a fictional team used as sample context for the workflows. Fork the repo and replace them with your own team's docs.

How We Work — team processes, meeting cadence, triage, and communication norms
Strategic Tradeoffs — the team's "even over" strategy statements, annotated with alignment evidence
Fake Google Docs Scope — synthetic external product, customer, security, launch, and program documents for folder- or shared-drive-based Google Docs demos
Launch Tracker Project — the sample GitHub Project with issues, launches, and workflow-generated artifacts
Intake Triage Project — project board for triaging incoming feature requests and bug reports
Customer Feedback Queue Project — project board for cross-channel customer feedback, dedupe clusters, and strategy/priority triage

Email: chrizbo@gmail.com
LinkedIn: linkedin.com/in/chrisbu

📄 License

MIT