From nimble
Researches meeting attendees and companies using real-time Nimble CLI web data. Surfaces roles, recent activity, context, talking points, and cross-attendee relationships.
npx claudepluginhub nimbleway/agent-skills --plugin nimbleThis skill is limited to using the following tools:
Research-powered meeting preparation with attendee intelligence and company context.
Generates structured pre-meeting briefing documents from company config, research files, Linear projects, meeting history, and fresh intelligence via researcher agent. Saves as Markdown file.
Generates sales call prep briefs with company/attendee research, agendas, and questions via web search and user input or CRM/email connectors.
Generates strategic sales meeting battle plans as self-contained HTML with coaching frameworks, belief stacks, scripted talk tracks, discovery questions, stakeholder maps, landmine warnings, and timed phase-by-phase game plans for discovery, demos, executive calls, and QBRs.
Share bugs, ideas, or general feedback.
Research-powered meeting preparation with attendee intelligence and company context.
User request: $ARGUMENTS
Before running any commands, read references/nimble-playbook.md for Claude Code
constraints (no shell state, no &/wait, sub-agent permissions, communication style).
Run the preflight pattern from references/nimble-playbook.md (5 simultaneous Bash
calls: date calc, today, CLI check, profile load, index.md load).
From the results:
references/profile-and-onboarding.md, stop~/.nimble/memory/people/index.md to identify existing
person profiles. Load relevant ~/.nimble/memory/people/ files for attendees
before — skip redundant searches, surface prior meeting notes. Follow
[[path/entity]] cross-references in person files: if an attendee's file links
to [[competitors/widgetco]], load that competitor file for richer context (e.g.,
recent intel from competitor-intel runs). Also check ~/.nimble/memory/companies/
for cached company research.
No same-day report check — meeting-prep is per-meeting, not per-day. Users
may prep for multiple meetings in one day. Instead, check entity freshness:
if a person/company profile was updated within the last 24 hours, offer to reuse
it: "I have a recent profile for [Name] from earlier today. Use it, or refresh?"Parse the meeting details from $ARGUMENTS or ask the user.
Calendar shortcut: If the user didn't specify attendees and a calendar connector
is available — either a calendar MCP tool (look for list_events in the tool list)
or the gws CLI (gws calendar +agenda --today) — offer to pull today's meetings
so they can pick one. If neither is available, skip this silently.
If clear (e.g., "prep me for my meeting with Alex Kim at WidgetCo tomorrow"):
If partial (e.g., "prep me for my meeting tomorrow"):
"Who are you meeting with? (names, titles, and company if you have them)"
If just a person (e.g., "research John Smith"):
Extract these fields:
| Field | Required | Source |
|---|---|---|
| Attendee name(s) | Yes | User input or calendar event |
| Company | Preferred | User input or inferred from search |
| Attendee title(s) | Optional | User input or discovered in Step 2 |
| Meeting type | Required | User input, inferred, or asked (discovery, demo, check-in, interview, partnership, internal) |
| Meeting date/time | Optional | User input |
| Additional context | Optional | User notes ("they're evaluating our product", "board member intro") |
Meeting type detection — if the user doesn't specify, infer from context clues:
| Signal | Inferred type |
|---|---|
| "prospect", "demo", "sales call" | Sales / discovery |
| "interview", "candidate" | Interview |
| "board", "investor" | Board / investor |
| "partner", "integration" | Partnership |
| "check-in", "sync", "1:1" with colleague | Internal |
| No signal | Ask (see below) |
If no signal — don't guess "general external." The meeting type gates whether the Value Positioning section is generated, so it's worth one question. Use AskUserQuestion:
What's the goal of this meeting?
- Sales / discovery — pitching, demo, exploring fit
- Partnership — integration, co-selling, joint venture
- Board / investor — board meeting, investor update, fundraising
- Interview — evaluating a candidate
- General / other — networking, catch-up, or not sure
Map the answer to the meeting type. If the user picks "General / other", treat as general external (no value positioning section).
The meeting type shapes the briefing focus — specifically, it determines whether the Value Positioning section (Step 4.5 + Step 6) is generated. Value positioning activates for: sales/discovery, partnership, board/investor. It is skipped for: interview, internal, general external.
Discover available WSAs for each attendee's company domain:
nimble agent list --search "{company-domain}" --limit 20
Run one search per unique company simultaneously. Filter for SERP/PDP WSAs,
prefer managed_by: "nimble", validate with nimble agent get --template-name {name}.
Cache discovered names + params. Pass them to attendee agents in Step 3 for richer
data. If no WSAs found, continue with nimble search alone.
Read references/attendee-agent-prompt.md for the full agent prompt template.
Follow the sub-agent spawning rules from references/nimble-playbook.md
(bypassPermissions, batch max 4, explicit Bash instruction, fallback on failure).
Check memory first. For each attendee, check ~/.nimble/memory/people/[name-slug].md.
If a profile exists and is < 30 days old, load it as known context and pass it to the
agent so it focuses on what's new. If > 30 days old, run a full refresh.
Spawn nimble-researcher agents (agents/nimble-researcher.md) with
mode: "bypassPermissions". One agent per attendee. Pass discovered WSA names
from Step 2 to each agent for enrichment.
Important: The Nimble API has a 10 req/sec rate limit per API key. With each agent running 4-5 searches, limit concurrent agents to 2 per batch. For 3+ attendees, batch in groups of 2.
Call estimation & Scaled Execution: Before launching agents, estimate total API
calls: ~5 searches per attendee + ~4 company searches + 3-5 extractions = ~(5 × N) + 9
calls. For 3+ attendees (15+ calls), tell agents to use extract-batch for page
extractions instead of individual calls. See the Scaled Execution pattern in
references/nimble-playbook.md for tier selection.
Batch 1 (2 agents simultaneously):
Batch 2 (if needed):
Single attendee optimization: If only one person, run the searches directly from the main context instead of spawning an agent — saves overhead.
Fallback: If any agent fails or returns empty, run those searches directly from the main context. Don't leave gaps in the briefing.
Before proceeding, verify every attendee has at least a title and company confirmed.
For any attendee with < 3 meaningful results or "Role Unknown":
--focus social fallback search directly (this searches social platform
people indices and is the most reliable way to find someone):
nimble search --query "[Name] [Company]" --focus social --max-results 5 --search-depth lite--focus social is unavailable, fall back to:
nimble search --query "[Name]" --include-domain '["linkedin.com"]' --max-results 5 --search-depth liteDo NOT present a briefing with "Role Unknown" — exhaust social search first. If still nothing after fallbacks, note it honestly: "Limited public presence — could not confirm role. Consider asking for their LinkedIn URL."
Also collect LinkedIn profile URLs for each attendee during this step if not already found. These are high-value for the final briefing output and Notion distribution.
Research the attendees' company for meeting-relevant context. This is a lighter version of company-deep-dive — focused on what's useful for the conversation, not a full 360°.
Company name quoting: If the company name contains common words that cause noisy
results (e.g., "Acme Supply", "Nova Dynamics", "Global Industries"), wrap it in escaped
quotes: "\"Acme Supply\" news". Use --include-domain '["[domain]"]' as an alternative anchor.
Make these Bash calls simultaneously:
nimble search --query "\"[Company]\" news" --focus news --start-date "[14-days-ago]" --max-results 8 --search-depth litenimble search --query "\"[Company]\" product launch OR announcement" --focus news --start-date "[14-days-ago]" --max-results 5 --search-depth litenimble search --query "about" --include-domain '["[domain]"]' --max-results 3 --search-depth litenimble search --query "\"[Company]\" funding OR raised OR investors" --max-results 5 --search-depth liteIf your user's company profile exists, also run:
nimble search --query "[Company] [UserCompany] OR [user-domain]" --max-results 5 --search-depth liteThis catches any existing relationship between the two companies — prior partnerships, mentions, shared investors, or competitive overlap.
If < 3 results from the news searches, retry without --start-date.
Date validation: When including company news in the briefing, verify that the
event date (when something actually happened) is recent, not just the article date.
See references/nimble-playbook.md → "Signal Date Validation" for details. If a snippet
uses past-tense language like "last year" or "back in Q3", treat it as background context
rather than recent news.
If the company was already researched (exists in ~/.nimble/memory/companies/),
load the existing profile and only run the news search for fresh updates.
Skip this step if the meeting type is interview, internal, or general external.
This step cross-references what you learned about the attendee's company (Step 4) with
the user's own business profile to find concrete positioning angles. It works best when
business-profile.json exists with at least company.name and company.domain.
If no profile exists, skip searches that reference the user's company or competitors (searches 2, 4, 5) and rely on generic research (searches 1, 3) for positioning insights. Use any WSAs discovered in Step 2 for richer attendee company data. The Value Positioning section will be thinner but still useful — pain-to-solution mapping and tech stack discovery work without a profile.
Load the user's sales context from ~/.nimble/business-profile.json:
sales_context.key_differentiators — what makes the user's product uniquesales_context.integration_partners — tools the user's product connects withsales_context.case_studies — similar customers and outcomessales_context.common_objections — pre-built objection responsescompetitors — tracked competitors (check if the attendee's company uses any)If sales_context doesn't exist in the profile, the skill still works — the value
positioning section will rely on web research alone rather than profile-enriched data.
Mention at the end: "Tip: Add sales context to your profile for richer positioning
next time."
Make these Bash calls simultaneously (3-5 searches depending on available data):
nimble search --query "\"[AttendeeCompany]\" tech stack OR tools OR platform OR uses" --max-results 5 --search-depth lite
→ Discover what tools/platforms they use — match against integration_partners
nimble search --query "\"[AttendeeCompany]\" [UserCompany] OR [user-domain]" --max-results 5 --search-depth lite
→ Any existing relationship, mentions, or competitive overlap (skip if already run in Step 3)
nimble search --query "\"[AttendeeCompany]\" challenges OR pain points OR struggling OR migrating" --max-results 5 --search-depth lite
→ Pain signals to map against user's value props
(If competitors list exists) nimble search --query "\"[AttendeeCompany]\" [CompetitorName1] OR [CompetitorName2]" --max-results 5 --search-depth lite
→ Check if they use a competitor — critical for displacement positioning
(If case_studies exist with matching industry) nimble search --query "[UserCompany] [attendee-industry] case study OR customer story" --max-results 5 --search-depth lite
→ Find published case studies in the attendee's industry to reference
From the results, extract:
This data feeds directly into the Value Positioning section in Step 5.
From Steps 3-4.5, identify the top 3-5 most informative URLs across all results. Prioritize:
Make one Bash call per URL, all simultaneously:
nimble extract --url "https://..." --format markdown
For extraction failures, follow the fallback in references/nimble-playbook.md.
Single attendee + known company: Skip company extraction, focus on person URLs. Multiple attendees: Prioritize person-specific URLs over company-level ones.
Structure the output as a meeting prep briefing. Adapt focus based on meeting type.
# Meeting Prep: [Company Name]
*[Meeting date/time if known] | Prepared [today's date]*
## Quick Take
[2-3 sentences: who you're meeting, why it matters, and the one thing to know
going in. This is the "read nothing else" paragraph.]
## Attendees
### [Name] — [Title]
**Background:** [Current role, time in position, career trajectory highlights]
**Recent Activity:** [What they've been posting, speaking about, or working on.
Direct quotes from posts/talks when available.]
**Conversation hooks:** [2-3 specific things to reference — shared connections,
their recent project, a post they wrote, a talk they gave]
**Notes from prior meetings:** [If exists in memory — what was discussed, their
preferences, open items. "No prior meetings on file" if none.]
[Repeat for each attendee]
## Relationship Map
[Cross-attendee connections — shared employers, mutual connections, overlapping
interests, organizational dynamics between attendees. Skip if single attendee.]
## Company Context
- **What they do:** [One line]
- **Size / Stage:** [Employees, funding stage, HQ]
- **Recent news:** [Top 2-3 items, dated with source]
- **Relevant to your meeting:** [How their company context connects to your
discussion — e.g., recent product launch you might discuss, funding that
signals growth, leadership change affecting priorities]
## Value Positioning
*[Only for sales/discovery, partnership, and board/investor meetings. Omit entirely
for interview, internal, and general external meetings.]*
### Value Mapping
[Match their specific needs/pain points to your capabilities. Every mapping must
be grounded in research from Step 4.5, not generic claims.
Format: "They [specific finding with source] → Your product [specific capability]"]
### Integration Hooks
[Tools/platforms they use that your product integrates with. Only include
integrations confirmed from research (their tech stack) AND your profile
(integration_partners). If no overlap found, say so honestly.]
### Recommended Positioning
[2-3 sentences on how to frame your pitch for THIS specific company and person.
Consider: their company stage, recent news, the attendee's role and priorities,
and any competitive displacement opportunity. This is the "elevator pitch
calibrated to this meeting" paragraph.]
### Reference Customers
[Similar companies from your case_studies that match their industry, size, or
use case. Include the outcome/metric if available. If no matching case studies,
omit this subsection rather than forcing a weak match.]
## Talking Points
[3-5 specific, actionable conversation starters grounded in the research.
Not generic "ask about their priorities" — specific: "Ask about their
migration from [old tool] to [new tool] that they announced last month."
When value positioning is active, weave 1-2 positioning angles into the
talking points naturally — don't make every talking point a sales pitch.]
## Watch Out For
[1-3 things to be aware of — sensitive topics (recent layoffs, bad press),
potential awkward overlaps, information gaps you couldn't fill.]
## Sources
[Numbered list of key URLs cited in the briefing]
Meeting type adaptations:
| Type | Emphasis | Add to briefing | Value Positioning |
|---|---|---|---|
| Sales / discovery | Buyer authority, pain signals, competitive stack | "Qualification signals" section | Yes — full section |
| Partnership | Mutual benefit signals, integration opportunities | "Alignment opportunities" section | Yes — focus on integration hooks |
| Board / investor | Financial context, market position, portfolio overlap | "Key metrics to reference" section | Yes — focus on recommended positioning |
| Interview | Candidate's work history depth, cultural signals | "Assessment angles" section | No |
| Internal | Skip company research, focus on person's recent work | Lighter format, no company section | No |
| General external | Balanced across all dimensions | Standard format above | No |
Core rules:
sales_context is missing from the profile, note it once at the end of the
Value Positioning section: "Tip: Edit your profile at
~/.nimble/business-profile.json to add sales context (differentiators,
integrations, case studies) for richer positioning next time."Make all Write calls simultaneously:
~/.nimble/memory/reports/meeting-prep-[company-slug]-[date].md~/.nimble/memory/people/[name-slug].md
(use the format in references/memory-and-distribution.md). Add [[path/entity]]
cross-references for the attendee's employer (e.g., [[competitors/widgetco]] or
[[companies/widgetco]]) and any other discovered relationships.~/.nimble/memory/companies/[company-slug].md if new
company data was found. Add reverse cross-references to the people researched
(e.g., [[people/alex-kim]]).last_runs.meeting-prep in ~/.nimble/business-profile.json
(only if profile exists)references/memory-and-distribution.md: update
index.md rows for all affected entity files, append a log.md entry for this run.The person profile in people/ should contain structured key facts (role, background,
interests, communication style) that can be loaded by future meeting prep runs.
Always offer distribution — do not skip this step. Follow
references/memory-and-distribution.md for connector detection, sharing flow, and
source links enforcement.
company-deep-dive for a full 360 on the company,
competitor-intel to track them as a competitor, competitor-positioning
to compare messaging before a sales meetingCheck at startup: echo $CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS
Team mode (flag set): Spawn teammates instead of sub-agents. Each teammate researches one attendee and can message the others when finding cross-connections.
| Teammate | Focus | Cross-checks with |
|---|---|---|
| Attendee 1 researcher | Full person + company research for attendee 1 | All other teammates (shared employers, connections) |
| Attendee 2 researcher | Full person + company research for attendee 2 | All other teammates |
| [Additional per attendee] | ... | ... |
How cross-attendee discovery works:
This produces higher-quality relationship maps than solo mode because teammates actively search for connections rather than just comparing results post-hoc.
Solo mode (flag not set): Standard sub-agent flow from Step 3.
competitor-intel for tracking competitorscompany-deep-dive for research without attendeesSee references/nimble-playbook.md for the standard error table. Skill-specific errors: