technographic-signals
Skill1 starsChanged 2 months agoMITView on GitHub
Find prospects by the technology they use, and use stack facts as outreach triggers. Use whenever the user mentions technographics, tech stack targeting, what tools a company uses, finding users of a competitor, BuiltWith or Wappalyzer style detection, integration directories, or wants to target companies on a specific platform. Also use when a product only fits companies with a particular tool in place.
In this repository30 other files it gives its agents.
- agency-multi-client-outboundSkill
- bounce-recoverySkill
- catch-all-risky-emailsSkill
- cold-email-copywritingSkill
- crm-data-hygieneSkill
- ecommerce-store-prospectingSkill
- email-deliverabilitySkill
- email-sequence-designSkill
- email-verificationSkill
- email-warmupSkill
- funding-news-signalsSkill
- google-maps-lead-scrapingSkill
- icp-definitionSkill
- inbox-rotation-scalingSkill
- job-posting-signalsSkill
- lead-enrichmentSkill
- lead-scoring-qualificationSkill
- lead-sourcingSkill
- linkedin-prospectingSkill
- meeting-booking-ctaSkill
- outbound-campaign-opsSkill
- outbound-complianceSkill
- outbound-metrics-benchmarksSkill
- personalization-at-scaleSkill
- reply-handlingSkill
- spam-filter-triggersSkill
- spf-dkim-dmarc-setupSkill
- subject-line-optimizationSkill
- suppression-list-managementSkill
- web-scraping-lead-dataSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
If your product integrates with, replaces or complements a specific tool, knowing which companies run it turns a broad market into a …
- Distinctive: line 29
Integration directories are the second: If a platform publishes a partner or app directory, the companies listed are confirmed users of …
- Distinctive: line 86
Don't cite low-confidence detections at all: Use them for segmentation, not for sentences.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Technographic Signals
If your product integrates with, replaces or complements a specific tool, knowing which companies run it turns a broad market into a qualified list. It's also one of the few filters that's genuinely predictive rather than merely descriptive.
The catch is accuracy. Detection methods vary from reliable to guesswork, and a list built on bad detection produces emails that confidently name the wrong tool — which is worse than not mentioning a tool at all.
Detection methods, by reliability
| Method | Reliability | What it finds |
| Job posts naming tools | Very high | The company stating its own stack |
| Public integration directories | Very high | Confirmed customers of a platform |
| Case studies and logos | Very high | Named, confirmed customers |
| Review site profiles | High | Verified users, with opinions |
| DNS and MX records | High | Email, hosting, CDN, security |
| Client-side script detection | Medium-high | Analytics, chat, tags, front-end tools |
| HTTP headers and cookies | Medium-high | Server, CDN, CMS, session tools |
| Careers or docs page mentions | High | Internal tooling |
| Aggregated technographic vendors | Medium | Broad coverage, dated and often stale |
| Inference from company profile | Low | Guesswork. Don't cite it in an email |
Job posts are the most reliable source and the most overlooked. A job description listing "Salesforce, Outreach, Gong required" is the company telling you its stack in writing, with a date. See job-posting-signals.
Integration directories are the second. If a platform publishes a partner or app directory, the companies listed are confirmed users of that platform. Same for case study pages and customer logo walls — those are curated lists of paying customers who agreed to be named.
What you can detect yourself
Front-end and infrastructure signals are visible in a page fetch and cost nothing.
# Front-end stack: analytics, chat, tag managers, frameworks
curl -s https://example.com \
| grep -ioE 'gtag|googletagmanager|hubspot|intercom|drift|segment|hotjar|klaviyo|shopify|_next|wp-content|webflow|marketo|pardot' \
| sort -u
# Server, CDN and platform
curl -sI https://example.com | grep -iE '^(server|x-powered-by|x-shopify|cf-ray|x-vercel|x-drupal)'
# Email provider from MX — reliable and stable
dig +short MX example.com
# Security and marketing tooling from SPF includes
dig +short TXT example.com | grep spfThe MX and SPF lookups are underrated. MX tells you Google Workspace versus Microsoft 365, which segments a market cleanly and never goes stale silently. SPF includes list every service authorised to send mail for the domain — CRM, marketing automation, transactional providers, helpdesk — which is a stack disclosure most people never think to read.
Front-end detection has real limits: it only sees client-side code, it misses anything server-side, and it can't tell an active deployment from a leftover script tag nobody removed.
The three plays
1. Competitor displacement. Companies using a competitor's product. The most valuable and the most crowded — they get this email constantly, so the message has to be specific about the actual difference rather than generic superiority. See what an honest version looks like in reply-handling under objections.
2. Complementary stack. Companies using a tool you integrate with or sit next to. Much easier to write, much less competitive, and the relevance is genuine rather than adversarial.
3. Missing capability. Companies that have adjacent tools but not yours — running paid ads with no analytics, or a store with no reviews widget. This is the strongest of the three when the gap is genuinely observable, because the pitch writes itself.
Using it in the email, carefully
Naming a company's internal tooling in a cold email is a fine line. Done well it proves relevance. Done badly it's presumptuous, and done wrong it's embarrassing.
Works — a fact they published, referenced neutrally:
Your job post for a RevOps lead lists Outreach and Gong.
Doesn't work — inferred, and stated as if you have inside knowledge:
I noticed you're using Salesforce.
The second reads as surveillance if it's right, and as incompetence if it's wrong. Prefer facts the company published about itself, and cite them in a way that makes the source obvious.
Never lead with a criticism of their stack. "I see you're still using X" insults whoever chose it, and that person is often the one reading the email.
Accuracy discipline
- Prefer stated over detected. A job post beats a script tag.
- Capture the source URL with every stack fact, so it's checkable.
- Note the detection date. Stacks change, and citing a tool they replaced last quarter is a bad opening.
- Allow the field to be empty. Forcing a detection produces a guess. See
personalization-at-scale. - Spot-check 20 by hand against the source before any campaign. Detection error rates are higher than vendors imply.
- Don't cite low-confidence detections at all. Use them for segmentation, not for sentences.
That last distinction is the practical rule: a medium-confidence signal is good enough to decide who gets the email, and not good enough to name in it.
Don't do these
- Don't buy a stale technographic list and quote it verbatim. Data is often a year old.
- Don't infer a stack from company size or industry and present it as observed.
- Don't claim inside knowledge. Reference the public source.
- Don't build the whole ICP on one detection. Combine with size, geography and a trigger.
- Don't scrape sources that prohibit it, or that require an account you'd be violating to use.
- Don't disparage the incumbent in a first email.
Sourcing it
The reliable sources — job posts, integration directories, case studies, docs, review profiles — are scattered across the web rather than sitting in one database, which is why most people fall back to a technographic vendor and inherit its staleness.
ManyPI can target this directly: keywords in a lead search expresses the technology signal, and extra_columns captures which tool was found, where it was stated and when, with the source URL. Because the agent reads live pages rather than a cached index, "mentioned in a job post last month" is a fact rather than an inference. Its scrapers can also be pointed at a specific integration directory and run on a schedule. Over MCP:
claude mcp add --transport http manypi https://mcp.manypi.com/mcp -s userRelated skills
job-posting-signals, lead-sourcing, lead-enrichment, personalization-at-scale, web-scraping-lead-data, icp-definition
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

