agentleFS
Intropix-teamIntropix-team/instagram-scraper-skills/skills/download-competitor-instagram/SKILL.md

download-competitor-instagram

Skill0 starsChanged 23 days agoMITView on GitHub

Download a competitor's Instagram (posts, reels, captions, engagement, music, locations, collaborations and sponsor brands) and audit it: what they post, how often, what performs, which brands and creators they work with. Use when the user wants to analyze, audit, benchmark or monitor a competitor's Instagram account, compare several brands' Instagram engagement, or export a brand's Instagram to a spreadsheet. Public business, creator and brand accounts; no Instagram login needed.

  • Installs packages

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Keep the results only as long as the user's purpose needs them (the terms suggest at most 30 days), and do not republish them as a dataset.

    line 18
  2. Very specific:

    If the user needs only dates, captions, likes, comments and views, the standard scraper intropix/instagram-posts-reels-scraper takes the …

    line 45
  3. Distinctive:

    Use the first path that is available.

    line 51
  4. Very specific:

    If the CLI says you are not logged in, ask the user to run apify login themselves (it opens a browser).

    line 78
  5. Very specific:

    If the run returned fewer posts than expected, read the run's OUTPUT record (apify key-value-stores get-value <defaultKeyValueStoreId> …

    line 119

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Download a competitor's Instagram

Runs the Intropix PRO Instagram Posts & Reels Scraper on Apify (intropix/instagram-posts-reels-scraper-pro), saves the posts, and writes a short audit of the account.

Rules

  • Public business, creator and brand accounts only. If the user asks for a private person's account, say the scraper's terms do not allow it and stop.
  • Tagged accounts, mentions and partners in the results can be private people: do not scrape them in turn or build profiles of them.
  • Keep the results only as long as the user's purpose needs them (the terms suggest at most 30 days), and do not republish them as a dataset.

1. Build the input

{"usernames": ["nike"], "maxPosts": 50, "includeOffGridReels": true}
FieldUse
usernamesThe competitor's handle, without @, or several to compare.
maxPostsHow many of the newest posts to read. 50 covers a few weeks for most brands.
maxPostsPerAccountInstead of maxPosts when comparing several accounts: up to this many of EACH account's newest posts.
sinceDate, untilDateOptional YYYY-MM-DD (UTC) bounds, e.g. last quarter.
includeOffGridReelstrue also counts reels shown only on the reels tab.
enrichPartnerstrue adds follower counts and categories for every co-author and sponsor brand ($0.005 per profile, paid Apify plans only).

Several competitors: one run with maxPostsPerAccount (e.g. 50) gives each account the same depth; split the results by username and compare. maxPosts alone is one shared total spent in input order, so the first account can use it all. A run holds at most 1,000 posts on a paid plan (10 on the free plan), filled account by account.

A free Apify plan delivers at most 10 posts per run and 40 results per day, enough for a first look; a real audit needs a paid plan.

If the user needs only dates, captions, likes, comments and views, the standard scraper intropix/instagram-posts-reels-scraper takes the same input at about half the price.

2. Run the scraper

Use the first path that is available.

A. Apify tools are connected (tools named call-actor and get-dataset-items, from the Apify connector or https://mcp.apify.com): call call-actor with actor intropix/instagram-posts-reels-scraper-pro and the input. If the run is not finished when the call returns, poll get-actor-run until it is. Then read the results with get-dataset-items using the run's dataset id, and set limit to the run's item count (the default is 20). With a shell, write the returned items, as one JSON array, to items.json for step 3.

B. A shell with the Apify CLI (apify --version prints 1.5 or later): write the input to input.json, then run

apify actors call intropix/instagram-posts-reels-scraper-pro --input-file input.json --json 2>/dev/null

Take storage.defaultDatasetId and storage.defaultKeyValueStoreId from the printed JSON (do not add --silent: together with --json it prints nothing). Then save the results:

apify datasets get-items <defaultDatasetId> --format json > items.json
apify datasets get-items <defaultDatasetId> --format csv > posts.csv

If the CLI says you are not logged in, ask the user to run apify login themselves (it opens a browser). Never ask the user to paste an API token into the conversation.

C. Neither is available: tell the user to either connect Apify (https://mcp.apify.com) or install the CLI (npm install -g apify-cli, then apify login). An Apify account is free to create. Stop there.

3. Save the media (if asked)

Media links expire within days. If the user wants the images and videos too, with a shell and Python 3:

python3 ${CLAUDE_SKILL_DIR}/scripts/save_media.py items.json instagram

(${CLAUDE_SKILL_DIR} is this skill's folder; use its real path if the variable is not set.) Files land in instagram/<username>/, each post recorded once in items.jsonl.

4. Write the audit

From the items, report in plain language:

  • Volume and rhythm: posts per week, the most common weekday and hour (UTC, from taken_at), share of reels vs photos vs carousels (post_type).
  • What performs: median and top engagement_rate, top 5 posts by engagement and by view_count with their permalink. Leave out posts with counts_hidden: true from rankings and say how many were left out.
  • Content: most used hashtags, recurring caption themes, licensed tracks vs original audio (audio_title, is_original_audio), tagged places (location_name, location_city).
  • Collaborations: coauthors, sponsor_usernames and posts with is_paid_partnership; with enrichPartners, the partners' follower counts from partners.
  • The account: followers, owner_category, owner_website, post count.

Null means "not available", never zero. For several competitors, put the key numbers side by side in one table.

If the run returned fewer posts than expected, read the run's OUTPUT record (apify key-value-stores get-value <defaultKeyValueStoreId> OUTPUT, or get-key-value-store-record with key OUTPUT) and explain its reason.

Cost

Paid by the user's own Apify account: $0.005 per run, $0.002 per account scanned, $0.0039 per delivered post ($3.90 per 1,000), plus $0.005 per enriched partner profile when enrichPartners is on. Failed lookups are not charged. Example: 50 posts of one competitor cost about $0.20.

Scraper page: https://apify.com/intropix/instagram-posts-reels-scraper-pro

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.