OpenClaw
Agents
Each agent has a job description, prompt, confidence threshold, blocked actions, required policies, and a human owner.
Executive
Leadership-facing synthesis and coordination agents.
Horizontal
Memory, governance, evaluation, and planning agents that serve every domain.
Vertical
Department-specific specialists for revenue, marketing, operations, CX, finance, and technology.
Registry Coverage
Canonical catalog-backed definitions now drive OpenClaw instead of scattered examples.
CEO Agent
Summarizes company health, KPI pressure, opportunity patterns, and strategic options for leadership.
Class: executive
Phase: Phase 4
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads cross-functional state, briefs, metrics, and risk posture. It never becomes final authority.
Allowed actions: summarize health, prepare executive brief, surface strategic opportunity
Required policies: executive-brief-policy, no-agent-action-without-command-gate
Chief of Staff Agent
Coordinates projects, tracks blockers, and follows up across departments.
Class: executive
Phase: Phase 4
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads tasks, approvals, dead letters, readiness, and active initiatives. It coordinates but does not hold legal authority.
Allowed actions: summarize blockers, create follow-up task, prepare weekly brief
Required policies: task-governance-policy, executive-brief-policy
Knowledge Agent
Maintains organizational memory across SOPs, meetings, documents, and durable operating truth.
Class: horizontal
Phase: Phase 1
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads Bravi Knowledge, structured records, linked tasks, and durable decisions. It may suggest structure but not mutate governed source truth silently.
Allowed actions: recommend knowledge link, create review task, assemble context pack
Required policies: knowledge-ownership-policy, knowledge-freshness-policy
Memory Agent
Promotes durable organizational memory instead of leaving important truth trapped in chat or notes.
Class: horizontal
Phase: Phase 1
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads continuity memory, contradictions, outcomes, beliefs, and source records.
Allowed actions: promote memory, flag contradiction, link decision evidence
Required policies: memory-promotion-policy, contradiction-visibility-policy
Evaluation Agent
Measures whether workflows, recommendations, and actions actually worked in reality.
Class: horizontal
Phase: Phase 2
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads outcomes, contradictions, audit runs, eval suites, and operating traces. It does not approve itself or override policy.
Allowed actions: summarize eval, recommend tuning, flag contradiction
Required policies: eval-integrity-policy, autonomy-promotion-policy
Governance Agent
Checks permissions, policies, risk, and compliance before any governed move advances.
Class: horizontal
Phase: Phase 1
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads policies, approvals, action intents, and role authority. It may route and explain, but it never bypasses policy law.
Allowed actions: evaluate action, route approval, explain block
Required policies: no-agent-action-without-command-gate, role-authority-policy
Learning Agent
Looks at failures and successes, then proposes changes to prompts, playbooks, and workflows.
Class: horizontal
Phase: Phase 2
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads outcomes, examples, prompts, and correction logs. It proposes changes but does not promote them directly.
Allowed actions: propose prompt refinement, summarize failure mode, recommend playbook change
Required policies: tuning-governance-policy, eval-integrity-policy
Planning Agent
Turns goals into projects, projects into tasks, and tasks into sequencing.
Class: horizontal
Phase: Phase 2
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads operating plans, task ledgers, readiness gaps, and linked knowledge. It may decompose work but not assign risky work outside authority rules.
Allowed actions: decompose initiative, suggest dependency, create planning draft
Required policies: task-governance-policy, founder-priority-policy
Marketing Agent
Owns campaign planning, copy suggestions, briefs, and testing ideas.
Class: vertical
Phase: Phase 2
Implementation: stub
Status: Dry Run
Owner: Irem
Jurisdiction: Reads campaign briefs, content workspace, experiments, and voice-of-customer signals.
Allowed actions: draft campaign plan, suggest brief, propose test
Required policies: campaign-launch-policy, content-approval-policy
Advertising Agent
Monitors Meta, Google, and LinkedIn ads and suggests optimization moves.
Class: vertical
Phase: Phase 2
Implementation: stub
Status: Inactive
Owner: Sultan
Jurisdiction: Reads paid metrics and campaign context. It never changes spend on its own.
Allowed actions: recommend budget shift, flag weak campaign, summarize attribution
Required policies: campaign-launch-policy, paid-spend-policy
SEO Agent
Tracks rankings, content gaps, and backlink opportunities.
Class: vertical
Phase: Phase 2
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads content library, search metrics, and ranking gaps.
Allowed actions: suggest seo update, flag content gap, recommend keyword cluster
Required policies: content-approval-policy, site-change-policy
Content Agent
Produces blog, LinkedIn, newsletter, and script drafts grounded in the brand system.
Class: vertical
Phase: Phase 2
Implementation: stub
Status: Dry Run
Owner: Irem
Jurisdiction: Reads campaign briefs, brand knowledge, files, and related tasks. It drafts but does not publish autonomously.
Allowed actions: draft post, draft newsletter, draft script
Required policies: content-approval-policy, brand-claims-policy
Research Agent
Researches competitors, industries, and trends to improve outbound and positioning.
Class: vertical
Phase: Phase 2
Implementation: mock
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads approved research inputs and internal strategy context. It does not act externally.
Allowed actions: draft research summary, compare competitor, flag trend
Required policies: research-source-policy
Lead Qualification Agent
Scores budget, timing, project fit, and intent to recommend the next governed sales move.
Class: vertical
Phase: Phase 1
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads lead, account, opportunity, task, and event context. It does not communicate externally.
Allowed actions: qualify lead, recommend next action, create internal follow-up task, log decision
Required policies: lead-response-policy, marketing-suppression-policy, no-agent-action-without-command-gate
CRM Agent
Maintains notes, reminders, and next actions so the CRM stays alive instead of decorative.
Class: vertical
Phase: Phase 1
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads leads, accounts, opportunities, activities, and tasks. It may propose updates but not mutate critical records without governed paths.
Allowed actions: draft note, suggest next action, create reminder
Required policies: crm-write-policy, record-linkage-policy
Proposal Agent
Creates proposals, quotes, and presentations grounded in approved economics and templates.
Class: vertical
Phase: Phase 1
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads opportunities, quote inputs, templates, and designer/project context.
Allowed actions: draft proposal, assemble quote context, flag missing economics
Required policies: proposal-gate-policy, margin-floor-policy, quote-profitability-policy
Meeting Prep Agent
Prepares history, objections, and opportunity context before a meeting.
Class: vertical
Phase: Phase 1
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads lead, account, opportunity, task, and communication history.
Allowed actions: prepare brief, suggest agenda, flag objection
Required policies: meeting-artifact-policy
Follow-up Agent
Creates personalized follow-ups, reminders, and response-aware next steps.
Class: vertical
Phase: Phase 1
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads conversations, tasks, cadence state, and meeting outcomes. It drafts but does not send autonomously.
Allowed actions: draft follow-up, create reminder, recommend next touch
Required policies: lead-response-policy, marketing-suppression-policy
Customer Success Agent
Monitors customer health, satisfaction, and renewal or escalation risk.
Class: vertical
Phase: Phase 3
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads issue history, order context, satisfaction signals, and service events.
Allowed actions: flag risk, suggest check-in, summarize service history
Required policies: customer-success-policy, refund-approval-policy
Support Agent
Answers questions and routes support issues safely into the right queue.
Class: vertical
Phase: Phase 3
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads support inbox items, case history, and policy. It drafts and routes but does not autonomously close sensitive issues.
Allowed actions: draft answer, classify issue, suggest escalation
Required policies: customer-issue-triage-policy, refund-approval-policy
Supplier Agent
Finds suppliers, pricing, lead times, and sourcing options.
Class: vertical
Phase: Phase 2
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads supplier docs, import data, and sourcing context. It does not contact suppliers autonomously.
Allowed actions: compare supplier, flag sourcing risk, recommend follow-up
Required policies: supplier-approval-policy, import-risk-policy
Inventory Agent
Tracks stock, shortages, and reorder timing.
Class: vertical
Phase: Phase 2
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads stock state, order demand, and shortage signals. It does not transact autonomously.
Allowed actions: flag shortage, suggest reorder timing, summarize inventory risk
Required policies: inventory-truth-policy
Scheduling Agent
Coordinates installations, meetings, and logistics timing.
Class: vertical
Phase: Phase 3
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads schedules, delivery readiness, installation context, and assigned owners.
Allowed actions: draft schedule plan, flag readiness mismatch, suggest next slot
Required policies: logistics-human-approval-policy, schedule-readiness-policy
Finance Agent
Tracks revenue, expenses, forecasts, and unit-economics pressure.
Class: vertical
Phase: Phase 3
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads finance records, quotes, refunds, and reconciliation state. It never finalizes money moves autonomously.
Allowed actions: summarize finance, flag anomaly, suggest forecast change
Required policies: finance-finality-policy, refund-approval-policy
Collections Agent
Handles invoices, reminders, and payment tracking for outstanding balances.
Class: vertical
Phase: Phase 3
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads invoices, payment status, and communication history. It drafts but does not send or threaten autonomously.
Allowed actions: draft reminder, classify delinquency, suggest escalation
Required policies: collections-communication-policy, finance-finality-policy
Recruiting Agent
Screens candidates and helps coordinate interview flow.
Class: vertical
Phase: Phase 4
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads role briefs, candidate notes, and interview signals. It does not hire autonomously.
Allowed actions: summarize candidate, suggest next step, prepare interviewer brief
Required policies: hiring-governance-policy
Training Agent
Teaches employees and answers internal company questions from approved knowledge.
Class: vertical
Phase: Phase 4
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads approved knowledge, SOPs, and training artifacts. It does not invent policy.
Allowed actions: assemble training path, answer internal question, draft onboarding checklist
Required policies: training-content-policy
QA Agent
Tests software, workflows, and outputs and reports issues with evidence.
Class: vertical
Phase: Phase 4
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads application surfaces, tests, and result logs. It reports; it does not ship.
Allowed actions: draft test case, flag regression, summarize bug
Required policies: qa-evidence-policy
Documentation Agent
Writes internal documentation from governed system knowledge.
Class: vertical
Phase: Phase 4
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads architecture, tasks, APIs, and knowledge items. It drafts but does not canonize without review.
Allowed actions: draft doc, summarize architecture, suggest missing doc
Required policies: knowledge-ownership-policy, documentation-review-policy
Sales Research Agent
Researches leads, accounts, and market context before outreach or discovery.
Class: vertical
Phase: Phase 2
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads lead, account, opportunity, and approved external research context. It never contacts prospects directly.
Allowed actions: prepare research brief, flag account signal, suggest research gap
Required policies: research-source-policy, crm-write-policy
Lead Scoring Agent
Scores lead quality, urgency, and fit using visible evidence rather than opaque vibes.
Class: vertical
Phase: Phase 1
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads lead profile, engagement history, source signals, and account context. It recommends, but does not suppress or disqualify autonomously.
Allowed actions: score lead, recommend priority, log scoring rationale
Required policies: lead-response-policy, marketing-suppression-policy
Outreach Draft Agent
Creates first-touch and follow-up outreach drafts for email-led cadences.
Class: vertical
Phase: Phase 1
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads lead context, account context, lists, cadence steps, and approved messaging guidance. It drafts, but does not send autonomously.
Allowed actions: draft outreach, suggest personalization, flag eligibility risk
Required policies: lead-response-policy, marketing-suppression-policy, no-agent-action-without-command-gate
Reply Classification Agent
Classifies prospect replies and recommends the next governed action.
Class: vertical
Phase: Phase 1
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads inbound replies, cadence history, and linked records. It classifies and suggests, but does not send or close records autonomously.
Allowed actions: classify reply, recommend next action, create follow-up task
Required policies: lead-response-policy, marketing-suppression-policy
Discovery Summary Agent
Turns discovery calls, notes, and uploaded transcripts into usable opportunity context.
Class: vertical
Phase: Phase 1
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads discovery notes, transcripts, files, lead state, and account context. It drafts structured summaries and follow-ups.
Allowed actions: draft summary, extract objections, recommend next step
Required policies: meeting-artifact-policy, crm-write-policy
Proposal Gate Agent
Checks whether a proposal or quote is actually ready to move forward.
Class: vertical
Phase: Phase 1
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads opportunity context, pricing completeness, proposal artifacts, and policy posture.
Allowed actions: assess proposal readiness, flag missing inputs, route approval
Required policies: proposal-gate-policy, margin-floor-policy, quote-profitability-policy
Quote Risk Agent
Reviews quotes for margin pressure, discount risk, and financial leakage.
Class: vertical
Phase: Phase 1
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads quote economics, discounts, shipping, duties, and linked opportunity context. It flags and routes; it does not finalize.
Allowed actions: flag margin risk, explain leakage, route approval
Required policies: margin-floor-policy, quote-profitability-policy, finance-finality-policy
Marketing Performance Agent
Reads campaign and channel performance and points to what should change next.
Class: vertical
Phase: Phase 2
Implementation: partial
Status: Active
Owner: Irem
Jurisdiction: Reads campaign metrics, channel data, and content performance. It recommends; it does not publish or change spend autonomously.
Allowed actions: summarize performance, flag underperformance, suggest test
Required policies: campaign-launch-policy, paid-spend-policy
Content Review Agent
Reviews content drafts for readiness, claims risk, and brand alignment.
Class: vertical
Phase: Phase 2
Implementation: partial
Status: Active
Owner: Irem
Jurisdiction: Reads content drafts, campaign context, and brand knowledge. It critiques, but does not publish or canonize.
Allowed actions: review content, flag risk, suggest revision
Required policies: content-approval-policy, brand-claims-policy
Social Sentiment Agent
Summarizes social and public sentiment patterns around campaigns, posts, and customer issues.
Class: vertical
Phase: Phase 2
Implementation: mock
Status: Dry Run
Owner: Irem
Jurisdiction: Reads social performance and public-response context. It never posts or replies autonomously.
Allowed actions: summarize sentiment, flag social risk, suggest review
Required policies: social-response-policy, content-approval-policy
CX Triage Agent
Classifies customer issues, drafts responses, and routes remedies through policy and approval law.
Class: vertical
Phase: Phase 3
Implementation: partial
Status: Active
Owner: Irem
Jurisdiction: Reads support context, order history, refund posture, and communication state. It drafts and routes, but does not finalize money or customer-facing promises autonomously.
Allowed actions: triage issue, draft response, recommend remedy
Required policies: customer-issue-triage-policy, refund-approval-policy
Logistics Packet Agent
Validates logistics handoff packets before work moves into execution.
Class: vertical
Phase: Phase 2
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads handoff packets, carrier requirements, approvals, and readiness evidence. It flags and routes; it does not book autonomously.
Allowed actions: validate packet, flag missing evidence, route approval
Required policies: freight-packet-completeness-policy, logistics-human-approval-policy
Finance Reconciliation Agent
Detects mismatches between quotes, invoices, payments, and refunds.
Class: vertical
Phase: Phase 3
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads finance records, operational truth, and reconciliation state. It flags drift, but does not finalize accounting truth.
Allowed actions: flag mismatch, summarize reconciliation issue, create follow-up task
Required policies: finance-finality-policy, refund-approval-policy, postgres-truth-policy
Email Performance Agent
Reviews draft, send, reply, and meeting patterns inside email-led sequences.
Class: vertical
Phase: Phase 2
Implementation: mock
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads email step metrics and outcome history. It does not send or alter sequence law autonomously.
Allowed actions: summarize metrics, flag weak step, suggest test
Required policies: lead-response-policy, email-telemetry-policy
Sequence Optimization Agent
Looks at cadence structure and recommends sequence improvements over time.
Class: vertical
Phase: Phase 2
Implementation: mock
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads enrollment history, reply classification, and performance summaries. It recommends changes but does not activate them.
Allowed actions: recommend cadence change, suggest branch, flag dead step
Required policies: tuning-governance-policy, lead-response-policy
Policy Guardian Agent
Specialized governance agent focused on explaining policy coverage, blocks, and escalation paths.
Class: horizontal
Phase: Phase 1
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads policy definitions, evaluations, approval records, and block histories. It advises but never overrides the law.
Allowed actions: explain policy block, summarize coverage, recommend remediation
Required policies: no-agent-action-without-command-gate, role-authority-policy
System Health Agent
Monitors connector freshness, projection drift, dead letters, and readiness posture across the system.
Class: horizontal
Phase: Phase 2
Implementation: partial
Status: Active
Owner: Sultan
Jurisdiction: Reads connector status, drift incidents, dead letters, readiness, and projection posture. It reports and routes, but does not clear production risks silently.
Allowed actions: flag drift, summarize readiness, create repair task
Required policies: postgres-truth-policy, worker-readback-policy, preflight-simulation-policy
Strategy Agent
Synthesizes trends, readiness, and opportunity signals into strategic options.
Class: executive
Phase: Phase 4
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads executive context, market research, and company performance. It proposes, not decides.
Allowed actions: draft strategic option, compare paths, summarize market pressure
Required policies: executive-brief-policy
Revenue Architecture Agent
Maps revenue systems, conversion pressure, and structural bottlenecks across marketing and sales.
Class: executive
Phase: Phase 4
Implementation: stub
Status: Dry Run
Owner: Sultan
Jurisdiction: Reads funnels, cadences, pipeline, and campaign performance. It recommends structure but does not override teams.
Allowed actions: summarize funnel, recommend system fix, flag bottleneck
Required policies: revenue-analytics-policy