Octave Claude Code Plugin

Access Octave’s GTM knowledge base directly from Claude Code for positioning, research, content generation, and sales enablement.

About Octave

Octave is a GTM intelligence platform that helps sales and marketing teams centralize positioning, messaging, and go-to-market knowledge. This Claude Code plugin brings Octave’s capabilities directly into your development workflow.

Installation

Prerequisites

• Claude Code CLI installed (install guide)
• Claude Code authenticated — a Claude subscription (Pro/Max/Team) or an Anthropic API key both work
• An Octave workspace account

Install via Marketplace (Recommended)

1. Add the Octave marketplace:

claude plugin marketplace add https://github.com/octavehq/lfgtm

2. Install the Octave plugin:

claude plugin install octave@lfgtm

3. Verify installation:

claude plugin list

You should see “octave” in your installed plugins.

Other AI tools (Codex & Cursor)

This repo is the source of truth. On every push to main, the same skills are automatically built and mirrored into editor-specific plugin repos:

• OpenAI Codex → octavehq/lfgtm-codex
• Cursor → octavehq/lfgtm-cursor

Install from the matching repo (see its README). Don’t edit those repos directly — they’re generated and overwritten. File issues and PRs here.

Quick Start

1. Configure MCP Server

Add the Octave MCP server (one connection per workspace):

claude mcp add octave-acme --transport http https://mcp.octavehq.com/mcp?ctx=

You can use any name that starts with octave- (e.g. octave-acme). Skills detect the Octave MCP server from your available tools.

Keep the ctx value private. It is a workspace-scoped credential — treat it like an API key and never commit it to a repository or paste it into shared documents.

2. Start Using Octave Skills

/octave:library list      # Browse your library
/octave:research [email protected] --for discovery  # Prep for a call

3. (Recommended for teams) Set Claude Org Preferences

If your team also has HubSpot, Salesforce, Gong, Granola, or Clay connected to Claude, set Claude Organization preferences so Claude routes GTM questions to Octave by default — reps won’t need to prefix prompts with “Using Octave, …” to get the right output.

See docs/org-instructions/ for short and long recommended instructions, admin setup, and test prompts.

Quick test (wait up to 1 hour for propagation first): in a new Claude conversation, ask “What’s the status of my deal with [company] and what should I be doing next?” — without mentioning Octave. If Claude reaches for get_deal or /octave:pipeline, the instructions are working.

Skills

Core Skills

Strategy & Messaging Skills

Research & Prospecting Skills

Sales Enablement Skills

Intelligence & Analytics Skills

Document Builder Skills

Library Operations

Publishing Skills

Agents

Specialized agent personas for sustained, multi-turn work sessions.

Skill Details

/octave:positioning

Complete visual Messaging & Positioning system as a scrollable HTML document:

• Message Framework (3-layer pyramid: market → product → value props by persona)
• Positioning Anchors (primary & secondary statements with keyword highlights)
• Positioning Strategy (tactical table: buyer, use case, problems, differentiators)
• Persona-Based Messaging (core message translated per buying committee role)
• Value Prop by Awareness Stage (4-column funnel: unaware → product-aware)
• Use Case Messaging Canvas (current way vs new way per use case)
• Use Case Lifecycle (customer journey phases with touchpoints)
• Homepage Messaging (website implementation: primary vs secondary messaging)

/octave:positioning                           # Full 8-section exercise
/octave:positioning message-framework         # Just the message framework
/octave:positioning anchors                   # Positioning anchors only
/octave:positioning homepage                  # Homepage messaging template
/octave:positioning --product "Platform"      # Focus on specific product

/octave:product-launch

Product and feature launch planning with content kit:

• Positioning and messaging by persona
• Channel strategy and timeline
• Content kit: announcement emails, blog, social, one-pager, FAQ, competitive talking points
• Library updates for new capabilities

/octave:product-launch "AI Analytics Dashboard" --type feature
/octave:product-launch "Enterprise Tier" --type product
/octave:product-launch "APAC Expansion" --type expansion

/octave:battlecard-doc

HTML or text competitive battlecards grounded in available evidence:

• Single-competitor deep dives with expandable sections and color-coded comparisons
• Full competitive landscape documents across every tracked competitor
• Diagnostic questions tied to the buyer’s decision criteria
• Objection preparation with attributed evidence and legitimate competitor advantages
• Displacement outreach offered as a follow-up

/octave:battlecard-doc --competitor "Acme"
/octave:battlecard-doc --competitor "all"   # Full competitive landscape doc

/octave:abm

Account-based planning:

• Deep account research and ICP scoring
• Stakeholder mapping with persona matching
• Per-stakeholder messaging strategy
• Engagement sequence recommendations
• Initial outreach generation

/octave:abm acme.com
/octave:abm acme.com --stakeholders 5 --motion "Enterprise Net New"
/octave:abm "Acme Corp" --depth quick

/octave:pipeline

Deal-level coaching:

• Stalled deal diagnosis and re-engagement strategy
• Multi-threading recommendations (find and engage more stakeholders)
• Competitive threat response
• Executive engagement strategy
• Closing tactics and expansion planning

/octave:pipeline stalled acme.com
/octave:pipeline multi-thread acme.com
/octave:pipeline competitive acme.com --competitor "Acme"
/octave:pipeline close acme.com

/octave:icp-refine

ICP analysis and refinement:

• Compare defined ICP against actual deal outcomes
• Identify winning patterns and anti-patterns
• Persona and segment effectiveness analysis
• Value prop resonance from conversation data
• Recommended library updates with evidence

/octave:icp-refine --period 90
/octave:icp-refine --segment "Enterprise"
/octave:icp-refine --focus losses

/octave:audit

Comprehensive library audit to identify:

• Missing or incomplete entities
• Orphaned personas / segments (not linked to any offering, so they don’t appear in any Motion matrix)
• Missing Motions for active offerings
• Stale content (not updated recently)
• Duplicate or overlapping entities
• Broken references
• Legacy standalone playbook → Motions migration

/octave:audit                    # Full audit
/octave:audit --type personas    # Focus on personas
/octave:audit --fix              # Interactive fix mode
/octave:audit --migrate          # Legacy playbook → Motions migration

/octave:prospector

Find and qualify companies and people matching your ICP:

• Searches using your library’s segments and personas as criteria
• Enriches, qualifies, and scores each prospect
• Provides filter suggestions for Apollo, Clay, or LinkedIn Sales Nav

/octave:prospector --motion "Enterprise Sales"
/octave:prospector --similar-to stripe.com
/octave:prospector --company acme.com

/octave:research

Context-aware research and prep:

• Discovery call prep (questions, pain points, qualification)
• Demo prep (use cases, proof points, objections)
• Outreach prep (hooks, angles, personalization)
• Pipeline review (deal health, next moves, risks)
• Recent Signals backed by verbatim call quotes with recording links, not paraphrases

/octave:research [email protected] --for discovery
/octave:research acme.com --for demo
/octave:research "Acme deal" --for pipeline-review

/octave:call-analyzer

Conversation analysis against your library:

• Resonance: Did messaging land?
• Adherence: Did we follow the Motion ICP narrative?
• Differentiation: Did we position effectively?

/octave:call-analyzer              # Paste content to analyze
/octave:call-analyzer --type call  # Analyze call transcript

/octave:insights

Surface intelligence from sales conversations:

• Top objections, pain points, and questions
• What’s resonating vs not
• Trends over time
• Library update suggestions
• Drill into the verbatim quote behind any finding, with recording link and timestamp

/octave:insights --type objections
/octave:insights --persona "CTO"

/octave:win-loss-report

Visual win/loss analysis as a self-contained HTML report with charts:

• Win/loss patterns and competitor analysis
• Period, segment, and competitor cuts
• Deal deep dives
• Verbatim objection and competitor quotes split by deal outcome, not paraphrases

/octave:win-loss-report
/octave:win-loss-report --competitor "Salesforce"
/octave:win-loss-report --period "2025" --segment "mid-market"

/octave:train

Practice selling with role-play simulations and knowledge quizzes:

/octave:train                                   # Interactive - pick a mode
/octave:train roleplay --persona "CTO"          # Role-play with a specific persona
/octave:train quiz --topic objections            # Quiz on objection handling
/octave:train quiz --competitor "Acme"           # Competitive knowledge check

/octave:ads

Build platform-ready ad campaign plans grounded in your library intelligence:

• Ad sets structured by persona, segment, or ICP
• 8 creative variant types: pain-focused, outcome, social proof, competitive, question-based, data-driven, status quo, authority
• Source cards with full derivation chains from library data to headline
• Audience targeting with positive/negative keywords and exclusions
• Landing page recommendations from your resources
• Export as CSV for Google Ads, Meta, or LinkedIn bulk upload
• Visual campaign deck as self-contained HTML

/octave:ads                                              # Interactive — full campaign builder
/octave:ads "compliance automation for VP Engineering"   # Campaign with angle
/octave:ads "competitive displacement vs Acme"           # Competitive campaign
/octave:ads "Q1 product launch"                          # Launch campaign

/octave:ads-resonance

Analyze ad performance and feed learnings back into your GTM library:

• Pulls performance data from an ads MCP server, BigQuery Data Transfer, direct API, or manual paste — auto-detected in that order
• Confidence-tiered findings that scale with spend, from smoke-test mode to full resonance analysis
• Falsifiable prediction cards scored on every run, accumulating a calibration track record over time
• Library update recommendations and a sales intelligence brief — applied only with your approval

/octave:ads-resonance         # Auto-detects the best available data source

/octave:deal-coach

Methodology-driven deal coaching built around Resonate → Elevate → Compel:

• Role play with stage-specific coaching agents and scored conversations
• Coaching microsites as self-contained HTML
• Coaching decks walking through the framework for a specific deal
• Interactive quizzes with deal-grounded scenarios
• Stage inference from CRM data, findings, and activity patterns
• Talking points grounded in the buyer’s own verbatim words, pulled across every call with the account

/octave:deal-coach                                        # Interactive — picks mode and stage
/octave:deal-coach acme.com                               # Ground coaching in a specific deal
/octave:deal-coach --mode roleplay                        # Jump to role play
/octave:deal-coach acme.com --mode microsite --stage compel  # Specific mode + stage
/octave:deal-coach --mode quiz --stage resonate           # Quiz on Resonate methodology
/octave:deal-coach acme.com --mode deck --stage elevate   # Coaching deck for Elevate

/octave:deck

Build Octave-powered HTML presentations with brand styling:

/octave:deck "pitch for Acme Corp"                          # Customer pitch
/octave:deck "Q1 QBR for enterprise segment"                # QBR with real data
/octave:deck --for competitive "vs Gong"                     # Competitive deck
/octave:deck ~/Downloads/existing-deck.pptx                  # Convert PPTX to HTML
/octave:deck "demo day pitch" --style octave-brand           # Specific style preset

/octave:champion-deal-room

Internal deal room a rep hands a champion so they can run the buying-committee sell:

• Quantified business case built from real deal data
• Stakeholder map with per-seat talking points
• Objection handling and a concrete path to yes
• Self-contained HTML — shareable inside the account

/octave:champion-deal-room acme.com
/octave:champion-deal-room acme.com --champion [email protected]

/octave:public-changelog

Publish entries to your workspace’s permanent public changelog. Paste what a staging page copied, or describe the items in plain words; the skill checks them, shows the exact lines being added, and publishes on your go.

/octave:public-changelog                  # Asks for the entries to publish
/octave:public-changelog     # Entries copied from a changelog staging page

The page is rendered by skills/public-changelog/scripts/changelog.py from a changelog.json stored beside it, never hand-edited: Octave re-reads it and treats every unseen line as a new capability, so a publish is refused unless every live line survives unchanged.

/octave:asset-manager

Publish and manage hosted assets on the Octave assets service:

• Cache-aware: lists existing assets before creating and offers matches (with links) so the same work isn’t done twice
• Upload local HTML sites, docs, or file bundles (interactive identifier + privacy intake — only_me / workspace / public)
• Update published files or metadata; move assets up or down the privacy ladder
• Create share links for specific emails or whole domains (never expire by default); add/remove recipients, revoke
• Persistent per-project registry of everything published (URLs, share links, status)

/octave:asset-manager publish ./use-cases-site   # Publish a folder
/octave:asset-manager share acme-use-cases       # Create/manage share links
/octave:asset-manager update acme-use-cases      # Replace files or metadata
/octave:asset-manager list                       # What have I published?

File uploads go through bundled curl scripts (skills/asset-manager/scripts/); metadata, shares, and tokens go through the asset_* MCP tools.

Configuration

MCP Server Setup

1. Add the Octave MCP server (one connection per project):

   claude mcp add octave-acme --transport http 

2. Skills detect the Octave MCP server from your available tools. No config file is required.

Authentication

Authentication is handled via OAuth at the MCP server level. No API keys are stored in the plugin configuration.

MCP Tools

The plugin uses the single Octave MCP server you configure (e.g. octave-acme). Call tools by name (e.g. verify_connection(), get_entity(...), list_entities(...)).

Connection

• verify_connection - Verify workspace connection and authentication status

Library Read

• list_entities - List entities by type; slim rows by default, includeDetails: true for full data, plus search, all, and pagination
• get_entity - Full entity details
• list_revisions - List version history for library entities
• get_revision - Get a specific historical revision of an entity
• search_knowledge_base - Semantic search
• ask_octave - Natural-language questions over the typed knowledge graph (entities, events, findings, opportunities)

Workspace Skills (managed)

Octave-hosted skill folders (Anthropic Agent Skills structure: root SKILL.md + optional bundled files) that teach agents reusable methods. Distinct from this plugin’s own Claude Code skills — these live in your Octave workspace and are retrieved by agents at generation time. Draft skills are invisible to agents until published. All members can read skills; create/update/delete require a workspace owner.

• find_skill - Resolve a task to the best-fitting published skill (returns name + description + confidence; empty result = proceed without a skill)
• list_skills - List skills (name + description only — progressive disclosure L1)
• get_skill - A skill’s SKILL.md instruction body + bundled-file listing (L2)
• get_skill_file - Fetch one bundled file’s content on demand (L3)
• create_skill - Create a skill: authored (name/description/body), import (URL or text → drafted SKILL.md), or from_examples (example outputs → drafted SKILL.md); lands as a draft
• update_skill - Update frontmatter/body/publish state (publishing makes it agent-discoverable)
• delete_skill - Delete a skill and its stored folder

Library Write

• create_entity - Create new entity - excludes legacy playbooks. Default: AI-generated from instructions. Verbatim: pass verbatim: true + fields and each value is stored exactly as written, no model involved — for approved copy the user does not want reworded
• update_entity - Update entity - excludes legacy playbooks. Rename deterministically with name / internalName, refine content with AI via instructions, or set specific fields to exact values with verbatim: true + fields (unnamed fields keep their value; a list you send replaces the stored list)
• get_entity_schema - Writable fields of an entity type (key, type, required, where it is stored, accepted nested keys) plus the workspace’s additional properties — call before a verbatim write to map the user’s copy onto the right keys. Verbatim writes accept fields.additionalProperties as a { key: value } map of those properties; an unknown key, nested key, or property fails the call with the allowed set listed, before anything is written
• delete_entity - Delete any entity type (soft delete)
• link_entities_to_offering - Link or unlink library entities (personas, segments, competitors, proof points, references, etc.) to a specific offering. Drives which entities appear in each Motion’s matrix.

Motions

• list_motions - List all Motions in the workspace
• get_motion - Full details for a Motion
• create_motion - Create a new Motion for an offering + motion type (NET_NEW, UPSELL, CROSS_SELL, CONVERT_FREE_TO_PAID, RENEW_AND_RETAIN, DISPLACE_INCUMBENT); auto-creates the Default Motion Playbook
• update_motion / delete_motion
• list_motion_playbooks / get_motion_playbook - Browse Motion Playbooks (Default + Custom) under a Motion
• create_motion_playbook - Create a Custom Motion Playbook with narrative type THEMATIC, MILESTONE, ACCOUNT, or COMPETITIVE
• update_motion_playbook / delete_motion_playbook
• list_motion_icps - List Motion ICP cells (persona × segment) for a Motion
• find_motion_icp - Full Motion ICP narrative (Target ICP overview, Operating landscape, Strategic narrative, Pains and consequences, Benefits and impacts, Methodology, References) plus optional Learning Loop learnings and Beats report context

Legacy Playbook Tools (deprecated)

Still available for workspaces operating on legacy standalone playbooks, but Motions and Motion Playbooks supersede them for new work. Use /octave:audit --migrate to translate.

• get_playbook - Get a legacy playbook with linked personas, segments, and value props
• list_value_props - List value props on a legacy playbook
• create_playbook / update_playbook - Manage legacy playbooks (avoid for new work)
• add_value_props / update_value_props - Manage value props on a legacy playbook

Configuration

• list_writing_styles - List all writing style configurations

Brand voices are retrieved via the generic entity tool: list_entities({ entityType: "brand_voice" }).

Resources

• list_resources - List global resources (documents, websites) with filtering
• get_resource - Get detailed resource information by oId
• create_resource - Create a new resource (text, file, URL, or Google Drive)
• delete_resource - Delete one or more resources
• search_resources - Semantic search across global resources

Research

• find_person / find_company - Search people/companies
• find_similar_people / find_similar_companies - Lookalike search
• enrich_person / enrich_company - Detailed intelligence
• qualify_person / qualify_company - ICP scoring
• resolve_profile_from_email / resolve_email_from_profile - Resolve a profile from an email (and vice versa)
• scrape_website - Scrape and extract structured content from a URL
• deep_web_research - Multi-source web research on a person, company, or topic

Content Generation

• generate_email - Generate email sequences
• generate_content - Generate various content types
• generate_call_prep - Generate call preparation materials

Events & Analytics

• list_events - Search calls, emails, deals
• list_findings - Aggregate extracted insights
• get_event_detail - Get detailed event info with transcript/content
• search_call_transcripts - Semantic + keyword search over indexed call transcripts. Returns verbatim, speaker-attributed moments grouped per call, with recordingUrl + startSec time anchors and live-hydrated linked CRM opportunities (current stage). Filterable by company, persona (attributedPersonaOIds, OR across the array), segment (attributedSegmentOIds, resolved live from company classifications), speaker side (customer vs. rep), deal outcome (WON/LOST/OPEN), sentiment, and date range. contentFilter narrows by exact phrase: momentPhrases (the quote itself must contain one), callPhrases (the call must mention one anywhere, e.g. “objections raised” + callPhrases: ["Loops"] = objections from calls that discussed Loops), and excludeCallPhrases (drop calls mentioning one). list_findings is the pipeline’s paraphrased insight; this is the raw conversation it came from — requires CAN_ANALYTICS and indexed transcripts (per-workspace backfill)
• get_entity_evidence - Best verbatim call quotes evidencing one library entity (persona, competitor, objection, use case, proof point). Prefers pipeline-linked moments, falls back to semantic search on the entity name — the entity-anchored composition of list_findings and search_call_transcripts

GTM Reports

Narrative GTM analyses (GTM Explorer / Beats).

• list_gtm_reports - List available GTM Explorer report groups
• get_latest_gtm_report - Get the most recent report (digest across configs in a group)
• get_report_run - Get the full content of a specific report run (title, summary, sections)

Suggestions

Proposed library changes (add / edit / merge) from conversation findings, queued for human review.

• list_suggestions - List entity suggestions (defaults to pending, last 14 days)
• get_suggestion - Preview a suggestion in full (proposed entity + current-vs-after diff)
• accept_suggestion / reject_suggestion - Apply or dismiss a pending suggestion
• create_suggestion - Queue a new pending suggestion (does NOT apply directly)
• update_suggestion - Revise a pending suggestion via natural-language instructions

Workspace Company

The workspace’s own company profile (singleton).

• get_workspace_company - Get the workspace company (null if not yet bootstrapped)
• update_workspace_company - Update workspace company fields (description, positioning, etc.)

CRM

• find_crm_records - Search for CRM records (accounts, contacts, leads, opportunities)
• find_crm_activities - Fetch activities (notes, tasks, calls, emails) for a CRM record
• generate_crm_context - Generate synthesized CRM context summary for a person or company
• get_crm_entity_schema - Introspect valid fields/properties on a CRM entity (discover field names before requesting them)

Pipeline Analytics

• list_deals - Paged synced-deal inventory: filter by open/stage/account/owner/pipeline/motion, sort by close date, amount or created date; stalledOnly for the at-risk portfolio (most severe first), includeRisk / includePulse per row
• get_deal - One deal by Octave oId or CRM id (Salesforce 15- and 18-character both work): core row plus opt-in activity, stakeholders, risk, pulse, history sections, each with a state saying why it is empty
• get_pipeline_metrics - Stage velocity, cycle time, win/loss conversion rates, and open deals per stage with summed value

Agents

• list_agents - List saved agents
• create_agent - Create a saved agent
• update_agent - Update a saved agent
• get_agent - Get a saved agent by oId
• delete_agent - Delete a saved agent
• run_email_agent - Run email sequence agent
• run_content_agent - Run content agent
• run_call_prep_agent - Run call prep agent
• run_enrich_person_agent - Run person enrichment agent
• run_enrich_company_agent - Run company enrichment agent
• run_qualify_person_agent - Run person qualification agent
• run_qualify_company_agent - Run company qualification agent

Assets

• asset_generate_access_token - Mint the per-user access token used by the upload/download scripts (rotates the previous one)
• asset_refresh_access_token - Rotate the access token (e.g. after a 401)
• assets_list - List your assets and your workspace’s assets, with status/privacy filters
• asset_get_by_id - Get one asset by uuid
• asset_update - Update metadata: identifier, description, entry point, privacy (only_me/workspace/public), status, vanity slug (pretty /s/// URL)
• asset_delete - Permanently delete an asset and its files
• asset_share_create - Create a share link for a non-public asset (emails and/or domains; never expires unless set; the returned URL is shown only once)
• asset_shares_list - List an asset’s share links
• asset_share_revoke - Revoke a share link (cuts active sessions)
• asset_share_add_recipients / asset_share_remove_recipients - Manage a share’s email allowlist
• asset_share_add_domains / asset_share_remove_domains - Manage a share’s domain allowlist
• asset_versions_list - List an asset’s immutable file versions (every file update mints one)
• asset_version_restore - Roll the served files back to a version (instant repoint; newer versions stay restorable)
• asset_version_delete - Delete an old version to free storage (the current version can’t be deleted)
• asset_stats_get - Per-day unique visit and download counts for an asset (owner-only)
• asset_visitors_list - Identified viewers: verified share recipients and workspace members (anonymous public visits are counted, never named)
• asset_access_requests_list - Cross-asset access-request inbox (filter by status; pending = needs action)
• asset_access_request_grant - Grant a request by minting a single-recipient share (the returned URL is shown only once; no email is sent)
• asset_access_request_dismiss - Dismiss a request (“not now” - the requester can re-request)

Directory Structure

.
├── .claude-plugin/
│   ├── plugin.json              # Plugin metadata (this repo root IS the octave plugin)
│   └── marketplace.json         # Marketplace config — publishes the root plugin as octave@lfgtm
├── .claude/
│   └── settings.json            # Project settings
├── .github/
│   └── workflows/               # CI: sync-downstream (Codex/Cursor mirrors), claude-tag zip
├── docs/
│   └── org-instructions/        # Recommended Claude org preferences for Octave-first routing
├── agents/
│   ├── asset-manager.md         # Hosted asset publishing/sharing agent
│   ├── octave-assistant.md      # General GTM assistant agent
│   ├── octave-editorial-reviewer.md     # Review-gate agent: language + information quality
│   ├── octave-presentation-reviewer.md  # Review-gate agent: visual + structural quality
│   ├── pmm-strategist.md        # Product marketing strategist agent
│   ├── sdr-coach.md             # SDR coaching agent
│   └── revenue-strategist.md    # Revenue strategy advisor agent
├── scripts/                     # Runtime helpers (resolved via ${CLAUDE_PLUGIN_ROOT}) + repo CI tooling
│   ├── deploy.sh                # Deploy a generated deck or page to Vercel
│   ├── export-pdf.sh            # Export an HTML presentation to PDF
│   ├── extract-pptx.py          # Extract content from .pptx for /octave:deck
│   ├── build-codex.sh           # CI: generate the Codex plugin artifact
│   └── build-cursor.sh          # CI: generate the Cursor plugin artifact
├── skills/                      # Skill definitions (each: SKILL.md + optional references/)
│   ├── shared/                  # Cross-skill reference layer (editorial & information rules, presentation principles, review protocol, entity model, style presets, formats/) — not a skill
│   ├── abm/                     # Account-based planning
│   ├── ads/                     # Ad campaign builder
│   ├── ads-resonance/           # Ad performance resonance loop + prediction cards
│   ├── asset-manager/           # Publish & manage hosted assets (bundled upload/download scripts)
│   ├── audit/                   # Library health check
│   ├── battlecard-doc/          # Interactive HTML competitive battlecard
│   ├── call-analyzer/           # Conversation analysis
│   ├── champion-deal-room/      # Champion-facing internal deal room (HTML)
│   ├── deal-coach/              # Deal coaching (Resonate/Elevate/Compel)
│   ├── deck/                    # Presentation builder (HTML)
│   ├── generate/                # Quick content generation
│   ├── get-brand-components/    # Brand kit capture for on-brand documents
│   ├── icp-refine/              # ICP refinement
│   ├── insights/                # Field intelligence
│   ├── library/                 # Library CRUD
│   ├── meeting-prep/            # Meeting battle plan (HTML)
│   ├── microsite/               # ABM microsite (HTML)
│   ├── one-pager/               # One-pager / leave-behind (HTML)
│   ├── pipeline/                # Deal coaching
│   ├── positioning/             # Positioning system (HTML)
│   ├── product-launch/          # Launch planning
│   ├── proposal/                # Business case / proposal (HTML)
│   ├── prospector/              # Prospect discovery
│   ├── qual-doctor/             # Qualification agent tuning
│   ├── research/                # Research & prep
│   ├── signals/                 # Morning intelligence briefing
│   ├── train/                   # Sales training & role-play
│   └── win-loss-report/         # Visual win/loss analysis (HTML)
├── EXAMPLES.md                  # Detailed usage examples
├── .gitignore
├── LICENSE
└── README.md

License

MIT - See LICENSE

Validation

See validation and release readiness for local checks, the behavioral evaluation matrix, and integration checks still required before claiming production verification.