spot-companies-using-specific-tech-stack（发现使用特定技术堆栈的公司）

Spot Companies Using Specific Tech Stack (Daily)

Spot companies using a specific tool, platform, or technology — inferred from their job postings. What a company requires in job posts tells you what they're running in production.

Technographic data alone is a trait (static). Combined with timing — "just adopted Snowflake" vs "has used Snowflake for 5 years" — it becomes an inferred pain-point and a leading indicator of buying behavior.

API Calls

This skill bundles two scripts in the same directory as this SKILL.md file. Never read or reference API credentials directly.

• signup.sh — handles authentication. Writes credentials to .env internally. Never exposes the API key.
• api.sh — handles all authenticated API calls. Reads credentials from .env internally.

First, resolve the script paths relative to this file's location:

SKILL_DIR="$(dirname "$(find ~/.agents/skills -name SKILL.md -path "*/spot-companies-using-specific-tech-stack/*" 2>/dev/null | head -1)")"
API="$SKILL_DIR/api.sh"
SIGNUP="$SKILL_DIR/signup.sh"

Then use $SIGNUP for auth and $API for all other calls.

When to Use This Skill

• "Find companies using Snowflake"
• "Find companies running Kubernetes in production"
• "Find companies using Terraform for infrastructure"
• "Find companies using dbt for data transformation"
• "Find companies with Kafka in their stack"
• "Find companies that recently adopted Datadog"

Important: Input Must Be Specific

Input must be a specific tool name — not general descriptions.

Good inputs: Kubernetes, Snowflake, React, Terraform, dbt, Kafka, Datadog, HubSpot Bad inputs: "cloud infrastructure" (too general), "data tools" (not specific), "modern stack" (meaningless)

Agent Rules

1. NEVER rewrite or reframe the user's query. Use the user's exact words as the search query. Do not add your own interpretation, expand abbreviations, add synonyms, or "improve" the query. If you think the query could be more specific, ask the user — do not modify it yourself.
2. Don't deploy signals without confirming. Signals cost credits. Always confirm before deploying.
3. Present what the API returns. No fabrication, no inference.
4. Reject vague inputs. If the user gives a general description instead of a specific tool, ask them to name the tool.
5. Never output or log API credentials. All authenticated calls go through api.sh.

Workflow

0. Agent Auth Check

Before anything, test if credentials are working by running:

bash "$API" POST /api/v1/signal/get-signal-list '{"pagination": {"limit": 1, "offset": 0}}'

If the call succeeds (returns JSON with signals): skip to Step 1.

If the call fails (returns an error or missing credentials message):

### Welcome to OpenFunnel

OpenFunnel turns daily events in your market into pipeline
— using OpenFunnel's Event Intelligence engine.

To get started, I'll authenticate you via the API.

**What's your work email?**

Wait for user input. Then:

1. Run bash "$SIGNUP" start "<user_email>"

- Returns {"status": "verification_code_sent", "email": "..."} on success

1. Tell the user a 6-digit code was sent: I sent a 6-digit verification code to **{email}**. Reply with the code.
2. Wait for input. Run bash "$SIGNUP" verify "<user_email>" "<code>"

- On success: returns {"status": "authenticated", "user_id": "..."}. Credentials are written to .env and .gitignore is updated automatically. - On failure: returns {"status": "failed",...}

1. Verify with bash "$API" POST /api/v1/signal/get-signal-list '{"pagination": {"limit": 1, "offset": 0}}'
2. If verification succeeds → continue to Step 1
3. If sign-up fails → ask user to retry
4. If verify fails → tell user the code was invalid or expired (up to 10 attempts in 24 hours), offer to retry or resend

1. Understand the request

What specific technology is the user looking for? If they give a vague description, ask for the specific tool name.

The deploy endpoint takes three fields:

• technographic_list — the primary tool names (e.g., ["Kubernetes", "K8s"])
• technographic_variations — common abbreviations, alternate names, related terms (e.g., ["k8s", "kubectl", "EKS", "AKS", "GKE"])
• technography_context — a short sentence explaining what you're looking for (e.g., "Companies running Kubernetes for container orchestration")

Help the user build these three fields from their input.

Timeframe: Last day to last year. Default: last 3 months. Shorter timeframes surface recent adopters. Longer timeframes surface established users.

2. Check existing signals

Run bash "$API" POST /api/v1/signal/get-signal-list '{"pagination": {"limit": 100, "offset": 0}}' to get all currently deployed signals.

A signal is unique by query + ICP pair. When checking for matches, compare BOTH:

1. Query match — close match on the technographic list.
2. ICP match — the signal's icp.id must match the user's intended ICP.

If potential match found (query + ICP both match):

I found an existing signal that covers this:

**{signal_name}** (ID: {signal_id})
**ICP:** {icp.name}

Want to use this one, or deploy a new signal?

If query matches but ICP is different:

I found a signal with a similar query but a different ICP:

**{signal_name}** (ID: {signal_id})
**ICP:** {icp.name}

This uses a different ICP than what you need. Want to:
1. Use this one anyway
2. Deploy a new signal with the right ICP

Wait for user input.

3. Get results from existing signal

Run bash "$API" POST /api/v1/signal/ '{"signal_id": <id>}' to get accounts and people matched by this signal.

### Results from: {signal_name}

**{total_accounts} accounts found | {total_people} people found**

If the user wants full details, run bash "$API" POST /api/v2/account/batch '{"account_ids": [<ids>]}'.

After presenting:

Would you like to:
1. See full details on specific accounts
2. Narrow results with filters (size, funding, location)
3. Deploy an additional signal for broader coverage ⚡ *uses credits*

4. ICP Check

Fetch available ICP profiles: bash "$API" GET /api/v1/icp/list.

If ICPs exist: present them and let the user pick one, or "none" to skip.

If the user types "none" or skips ICP selection:

Auto-create a broad fallback ICP:

bash "$API" POST /api/v1/icp/create '{"name": "Broad Default ICP", "target_roles": ["Any"], "employee_ranges": ["1-10", "11-50", "51-200", "201-500", "501-1000", "1001-5000", "5001-10000", "10001+"], "location": ["Any"]}'

No ICP selected, so I created a broad fallback ICP: **{name}** (ID: {id})

Using this ICP for your signal.

If no ICPs exist:

You don't have an ICP profile yet. A quick one will make results much sharper —
it filters by company size, location, and the roles you're targeting.

1. **Quick setup** (recommended) — takes 30 seconds
2. **Skip** — auto-create a broad fallback ICP and continue

If quick setup → collect ICP name, target roles, company size, and location. Create via bash "$API" POST /api/v1/icp/create '...'.

If skip → auto-create the broad fallback ICP as above.

5. Confirm & Deploy

I'll deploy a **technography** signal:

**Name:** {auto-generated descriptive name}
**Tech:** {technographic_list}
**Variations:** {technographic_variations}
**Context:** {technography_context}
**Timeframe:** {default — 90 days}

**ICP:** {selected or created ICP name}

⚡ *This will use credits from your plan.*

Other options:
- **Repeat daily** — re-run this signal every day for continuous monitoring
- **Audience name** — auto-add results to a named audience
- **Credit limit** — cap spending on this signal

Set any of these, or "deploy" to go with defaults.

Wait for user input. Then deploy:

bash "$API" POST /api/v1/signal/deploy/technography-search-agent '{"name": "<name>", "technographic_list": ["<tech>"], "technographic_variations": ["<variations>"], "technography_context": "<context>", "timeframe": <days>, "icp_id": <id>, "repeat": <true|false>}'

6. Post-deploy

Signal deployed: **{name}** (ID: {signal_id})

This is now scanning job posts for companies using this technology.
Results come in as they're found — just say "check on {signal_name}" anytime.

7. Check back

Run bash "$API" POST /api/v1/signal/ '{"signal_id": <id>}' to get results found so far.