Insta Parcel Skill

An Agent Skill for Claude Code and Codex that turns public or permissioned Instagram creator research into an auditable influencer lead list.

It guides an agent through creator discovery, profile vetting, follower thresholds, engagement-rate (ER) calculation, location or audience evidence, CSV normalization, deduplication, human review, and safe publication.

Install

Claude Code

npx --yes skills add rdsolod-ui/insta-parcel-skill \
  --skill insta-parcel-skill \
  --agent claude-code \
  --global \
  --copy \
  --yes

Invoke with:

/insta-parcel-skill

Codex CLI

npx --yes skills add rdsolod-ui/insta-parcel-skill \
  --skill insta-parcel-skill \
  --agent codex \
  --global \
  --copy \
  --yes

Invoke with:

Use $insta-parcel-skill to validate this creator research batch.

What It Handles

• Instagram creator discovery from public or permissioned sources
• influencer research and creator prospecting
• public profile vetting and evidence review
• follower thresholds and account-type rules
• engagement-rate calculation from visible likes and comments
• creator lead-list CSV validation
• handle normalization and duplicate detection
• manual-review gates before outreach or publication
• additive dataset previews, quarantine, and release checks

The skill is designed for marketing teams, influencer marketing workflows, creator economy research, social media analytics, and reviewable lead generation. It does not ship a covert scraper or bypass platform controls.

Example Prompts

Find public Instagram creators for an influencer marketing campaign and keep
an evidence trail for every candidate.

Review this creator CSV, calculate ER, reject duplicates, and flag profiles
that need manual verification.

Build an Instagram influencer lead list from permissioned sources without
using private APIs or bypassing login and rate limits.

Validate this batch before it is merged into our creator database.

Architecture

skills/insta-parcel-skill/
├── SKILL.md
├── agents/
│   └── openai.yaml
├── references/
│   ├── data-contract.md
│   ├── evidence-policy.md
│   └── publish-checklist.md
└── scripts/
    └── validate_batch.py

The main instructions stay concise. Detailed policy, schema, and publication controls load only when the active task needs them. The bundled validator is dependency-free and provides deterministic checks that should not be left to model judgment.

Validate A Batch

python3 skills/insta-parcel-skill/scripts/validate_batch.py \
  examples/sample-batch.csv \
  --min-followers 3000 \
  --allowed-account-type personal \
  --require-er \
  --require-evidence

Machine-readable report:

python3 skills/insta-parcel-skill/scripts/validate_batch.py \
  examples/sample-batch.csv \
  --min-followers 3000 \
  --require-er \
  --require-evidence \
  --json

Safety Boundary

Use public data, official exports, approved APIs, or a browser session where the user logged in manually. Stop at login prompts, CAPTCHA, checkpoints, or rate limits.

Do not use this skill for:

• private-profile collection;
• sensitive-trait inference;
• covert surveillance;
• anti-abuse bypass or proxy rotation;
• automated likes, follows, comments, messages, or account changes;
• unreviewed outreach lists;
• collecting or publishing data about minors.

Review the full evidence policy before adapting the workflow.

Discovery And Quality

The repository deliberately uses precise language rather than keyword stuffing. See Discovery and quality for:

• how Claude and Codex select skills;
• how skills.sh search and ranking work;
• the trigger-query test set;
• the trust signals maintainers should improve over time.

Development

python3 -m unittest discover -s tests -v
python3 tools/validate_skill.py skills/insta-parcel-skill

Contributions are welcome. Read CONTRIBUTING.md and SECURITY.md before opening a pull request.

Non-Affiliation

This project is not affiliated with, endorsed by, or sponsored by Instagram or Meta. Users are responsible for complying with applicable platform terms, privacy requirements, and local law.