influencer-discovery skill
A skill that searches 15 channels for creators with their own audience, gathers public contacts and appends them to a Google Sheet
- 222 ★ on the repository
- Last commit
Broad access
Why this level
- Collects people's personal contacts (names, emails, profiles) and writes them into a Google Sheet
- Calls platform APIs and sites with your keys and sessions, and the sheet token is minted through a service account
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/influencer-discoveryDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Collects people's personal contacts (names, emails, profiles) and writes them into a Google Sheet
- Calls platform APIs and sites with your keys and sessions, and the sheet token is minted through a service account
This is third-party code. Review the repository files before installing.
What it does
influencer-discovery is a creator-discovery pipeline that runs as a Claude Code skill. It looks for people with their own audience (followers, readers, listeners), pulls their public contacts and appends rows to a Google Sheet. It filters out people building their own product: they want reach and will not hand it out. Channels are split into tiers. Tier one is social: X, Instagram, TikTok, Threads, YouTube, Reddit, Mastodon. Tier two is blog platforms DEV.to, Hashnode, WordPress.com and Micro.blog. Tier three is personal sites, newsletters and podcasts, and tier five is the media outlets freeCodeCamp News and HackerNoon. For each channel `reference/datalayer/` documents what its API exposes and which credentials it needs, and `reference/methodology/` explains how contacts are obtained. The cooperation priority is set by the folder name prefix. A run starts by hand with `python3 skills/influencer-discovery/scripts/run.py` and the flags `--tiers`, `--channels`, `--subject`, `--min-followers` and `--min-karma`. Writes to the sheet only append rows, never edit existing cells, and abort if the header does not match. The skill stops at capturing contacts: sending messages, deal tracking and settlement are out of scope.
Who it is for. Marketers and founders who look for creators to promote a product and build a contact list in a sheet instead of working through each platform by hand.
Good fit when
- You need a batch of creators in a niche, each with an email or other contact
- A candidate sheet has to be topped up regularly, without duplicates or repeat entries of the same people
- The list has to be narrowed by audience with `--min-followers` and `--min-karma`
- Several platforms fit at once: from Reddit and YouTube to newsletters and podcasts
- You are working out how to get an author's email on a specific platform: the method for each channel sits in `reference/methodology/`
Not a fit when
- You need outreach or deal tracking: the skill stops at capturing contacts
- Collecting personal contacts without consent is not allowed under your country's law
- You have no Google Sheets or a service account with edit rights on the sheet
Example request
Find 20 Rust development bloggers with audiences of 5000 followers or more, pull their public contacts and append them to the target sheetLimitations
- The skill gathers people's personal contacts. Obligations under personal data law (GDPR and similar) stay with you.
- Some channels work without credentials, while the rest need an API key or a logged-in account; per-channel requirements are in `reference/datalayer/index.md`.
- Writing to Google Sheets uses a short-lived token from a service account, which needs the sheets, iam and iamcredentials APIs enabled and the serviceAccountTokenCreator role.
- Linux and macOS are supported and Python 3.11 or newer is required. The LinkedIn channel sits in the `_not-run` folder.
How to disable. Delete the skill folder from the agent's skills directory or the repository clone. Credentials live in `~/.config/influencer-discovery/.env` and pipeline state in `~/.local/share/influencer-discovery/`; both folders are removed by hand. Delete rows in the sheet separately.
FAQ
How do I install influencer-discovery?
Clone the repository and run `python3 skills/influencer-discovery/scripts/run.py --help`, or copy the `skills/influencer-discovery/` folder into the agent's skills directory. No third-party packages are needed, just the Python 3.11 standard library.
Which creators count as a match?
Those who bring their own audience: followers, readers, subscribers. Authors building their own product are filtered out as false positives.
Does the skill edit existing sheet rows?
No. Writes only append rows and abort if the header does not match. Duplicates are caught by the person-and-platform key against a local log.
How does the skill treat anti-bot protection?
Requests are polite: an explicit User-Agent, throttling and no anti-bot circumvention. All external requests leave through a single read-only choke point.
Does the skill send emails to the people it finds?
No. The skill stops at capturing contacts. Outreach, deal tracking and settlements with creators are out of its scope.
Discussion
No questions or reviews yet.
Related
SkillsSix skills for Codex run TikTok product selection, account audits, lead-generation content and ads reviews, driven by data, not guesswork.
A Claude Code workspace that tracks jobs, finds roles and sends first messages to 1st-degree LinkedIn connections
A local-first set of Codex skills and a read-only CLI that turns your own WeChat history into search, digests, follow-ups and opportunity tracking
TKBK Skills Hub: skills for marketplaces, CRM and Avito
TKBK Skills Hub
A skill collection where ten own skills cover Wildberries, Ozon, Yandex Market, Bitrix24, amoCRM, Avito and YooKassa, plus imported ones
A skill that connects two Avito MCP servers: the official API for your own account and public search for competitor research
GTM Skills by Explorium
GTM Skills
22 skills for B2B sales: company lists, contact enrichment, lead scoring and cold email, built on Vibe Prospecting MCP data
README in short
The README presents the project as a creator discovery pipeline that runs as a Claude Code skill and lists channels by tier. It then covers install with no external packages, a sample run with flags, Google Sheets access through a service account and a table of principles: a dedup key, append-only writes, a one-record blast radius, distrust of page content and keeping contacts out of the repository.
README badge
Original SKILL.md
--- name: influencer-discovery description: > Find the bloggers/creators who can help promote your work, capture their contact info, and append them to the target sheet in Google Sheets. Targets must bring their own audience — followers, readers, subscribers; people building their own product are not targets. Use for requests like "find a batch of bloggers doing X", "top up the target sheet", "dig up contact info for these people", "how do I get emails on this platform". Covers X/Twitter, TikTok, Instagram, Threads, YouTube, Reddit, Mastodon, DEV.to, Hashnode, Blog, Newsletter, Podcast, plus distribution Websites (tool-recommendation sites, directory sites). Stops at capturing contact info. Sending, tiering, deal tracking, and click settlement are all out of scope — do not use this skill for those requests. --- # influencer-discovery One sentence: **hit the qualified-row count the user asked for, every row with contact info, not one row a duplicate.** ## Who to find **People who can help us promote our work.** The criterion is **bring your own audience**: followers, readers, subscribers — when they post something, someone sees it. **People building their own product are not targets.** They want traffic and won't distribute for us — capturing their email is a false positive, not output. This is the biggest false-positive source in this methodology; see [seller-vs-buyer.md](reference/methodology/_shared/seller-vs-buyer.md) for how to tell. **The publication itself is not a target; its authors may be.** Multi-author publications have editors and pages, and the site only exposes a submissions inbox — one row per site collapses hundreds of authors into a single row. The move is to pull the author list from the publication and go out through the authors; see [5-media/](reference/methodology/5-media/index.md). ## Read these first Before starting, read `docs/design/index.md`, then the docs for the areas involved. **Invariants and scope are authoritative there; this file does not restate them**: - Append-only never modify, the log records everyone while the sheet records only the qualified, failures are not logged, better to miss a row than fabricate one — [index.md](../../docs/design/index.md) - Two shapes and three stop semantics, the two hops, fetching rules — [platforms.md](../../docs/design/platforms.md) - What the log / target sheet / run report each record — [records.md](../../docs/design/records.md) - Dedup rules and their premise — [dedup.md](../../docs/design/dedup.md)