One responsibility with a clear result and boundary.
StackOS Library / agents
Specialist agents with clear jobs.
Each agent has one focused responsibility inside a larger workflow—from research and creation to review, safety, and verification.
Quick answer
What are AI agents for business?
AI agents for business are focused roles that use a model, instructions, context, and connected tools to complete a clear responsibility. StackOS places those specialists inside complete workflows, where their work can be understood, reviewed, and handed to the next stage.
Compare agents, workflows, and orchestrators →Works through Codex, Claude Code, Gemini, and other clients.
Uses the apps and accounts your team already relies on.
Architecture Agent
Designs implementation structure, ownership boundaries, schemas, operation contracts, and one-brain reuse before delivery.
Branding Channel Strategist
Owns angle, structure, and channel publication job decisions without executing publication.
Branding Claim Auditor
Read-only adversarial reviewer that enforces evidence-over-assertion.
Branding Evidence Curator
Interviews, researches, harvests, classifies, dedupes, and captures operational receipts as traceable evidence-items.
Branding Narrative Writer
Drafts main pieces and channel renditions from cleared evidence in the project-adapted voice.
Branding Sanitization Reviewer
Read-only disclosure reviewer that fails closed with cleared or blocked verdicts.
Branding Voice Reviewer
Read-only reviewer for base voice, channel register delta, form contract, and kill tests.
Campaign Brief Analyst
Converts raw person inputs into a complete, professional campaign brief resource with persisted input artifacts and a proposed campaign workspace slug.
Campaign Creative Director
Turns the approved brief into the campaign plan whose media manifest, page plans, and acceptance criteria form the traceability contract for production and signoff.
Campaign Landing Page Builder
Builds each planned landing page variant as a self-contained static HTML file in the campaign workspace and assembles gallery.html as the person-facing index.
Campaign Media Producer
Executes approved image manifest entries through granted generation actions with drift control, set consistency, cost discipline, and truthful per-asset records; video requests are…
Campaign Visual Signoff Reviewer
Verifies every produced asset and page against the approved campaign plan's acceptance criteria with rendered evidence and per-item verdicts instead of rubber stamps.
Codebase Explorer
Read-only mapper for affected scopes, work paths, ownership boundaries, and downstream fallout before code changes.
Communications Callback Follow-Up Agent
Handles callback or follow-up requests with origin context, route policy, and reply/send safety.
Communications Customer Feedback Intake Agent
Normalizes customer feedback from Slack, Telegram, or another channel into one route-approved main Slack support thread while preserving media and handoff refs.
Communications Inbox Review Agent
Reviews stored inbound communications, triages requests, and prepares safe workflow or reply handoffs.
Communications Outbound Notification Agent
Prepares and sends explicit connected app-neutral outbound notifications through named targets.
Communications Rich Telegram Reply Agent
Sends or validates Telegram replies with rich features while preserving exact delivery semantics.
Delivery Agent
Implements tickets one by one, debugs root causes when checks fail, updates tracker state, verifies behavior, and preserves project conventions.
GTM Account Research Agent
Produces account research briefs with source clarity, ICP fit, risks, and outreach-ready handoff.
GTM CRM Hygiene Agent
Plans and reviews CRM cleanup with field-level atomic updates, duplicates, and audit visibility.
GTM Lead Enrichment Scoring Agent
Enriches and scores leads according to project-specific fields, confidence, and review policy.
GTM Outbound Sequence Agent
Prepares outbound sequence content and targeting with research, compliance, and approval context.
GTM Pipeline Risk Review Agent
Reviews pipeline risk, stale opportunities, and next actions using CRM context and project-specific stage rules.
Landing Page Creative Experiment Agent
Coordinates landing-page and creative experiment planning across UX, media, tracking, and review.
Media Budget Reallocation Agent
Reviews budget movement proposals with evidence, constraints, and approval awareness.
Media Campaign Launch Agent
Prepares paid media campaign launches with explicit assumptions, approvals, and connected app-action boundaries.
Media Creative Variant Agent
Generates and reviews creative variant plans tied to hypothesis, brand constraints, and measurement.
Media Performance Diagnosis Agent
Diagnoses media performance using bounded evidence, live-data caveats, and prioritized next checks.
Planning Agent
Turns scoped work into tracker tasks, dependencies, ticket slices, acceptance criteria, and verification strategy.
Project Memory Review Agent
Reviews bounded StackOS project memory and prepares a grounded next-step recommendation.
Release/Ops Agent
Handles release readiness, rollout notes, post-release checks, and operational cleanup.
Requirements Flow Definer
Turns broad intent into explicit project-adapted user, data, system, and business flows, acceptance criteria, non-goals, and evidence expectations before design begins.
Review Agent
Reviews designs and completed delivery across behavior, flow-regression coverage, QA, integration contracts, security, tracker evidence, and release risk.
SEO Content Refresh Agent
Reviews stale or underperforming content and prepares a refresh plan with source-backed changes.
SEO Keyword Research Agent
Plans and reviews keyword discovery workflows with evidence, constraints, and content opportunity handoff.
Support Delivery Handoff Agent
Converts a support conclusion and same-thread person instruction into delivery-ready tracker work with durable chat refs.
Support Issue Investigator
Investigates the full main support thread and project evidence, asks bounded clarification questions, and posts a thread-bound conclusion without implementing fixes.
Test Designer
Maps flow design, requirements, and design risks to focused tests, proof surfaces, TDD gates, agent-executed E2E/manual proof depth, and signoff evidence.
Trackbooth Agent API person
Operates Trackbooth Agent API inventory only through StackOS manual catalog sync, generated actions, and StackOS-held API-key auth.
Workflow Author
Designs reusable StackOS workflow contracts, project workflow extensions, and concrete run plans from project-safe context.