agentleFS
Sign inStart free
TomGranotTomGranot/hubspot-admin-skills/CLAUDE.md

hubspot-admin-skills

CLAUDE.md75 starsChanged 20 days agoMITView on GitHub

This is a Claude Code skills marketplace plugin for HubSpot CRM administration.

  • Reads credentials

We split this CLAUDE.md into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    All content is company-agnostic — no customer data, credentials, or proprietary references

    line 11
  2. Very specific:

    Before opening a PR: claude plugin validate .

    line 14

≈ “Common” is an estimate: at least 1 in 5 of the 25,680 CLAUDE.md 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
CLAUDE.md

HubSpot Admin Skills

This is a Claude Code skills marketplace plugin for HubSpot CRM administration.

Conventions

  • Each skill lives in skills/<skill-name>/SKILL.md; skills with scripted stages ship scripts/before.py, scripts/execute.py, scripts/after.py
  • Skills follow the Agent Skills Specification: context, prerequisites, step-by-step instructions, verification checks, and rollback guidance
  • Skills that change portal state follow the Plan → Before → Execute → After pattern with a required Rollback section, an abort threshold, a CSV audit trail, and explicit user confirmation before mutating anything
  • The repo is organized by CRM lifecycle phase: audit, hygiene, enrichment, segmentation, automation, and ongoing maintenance
  • All content is company-agnostic — no customer data, credentials, or proprietary references
  • Scripts are plain Python using requests against HubSpot REST endpoints directly (no SDK wrapper), with PEP 723 inline metadata, run via uv run from the user's project directory
  • Inside a SKILL.md, script paths are relative to the skill's own directory (scripts/before.py, or ../other-skill/scripts/x.py for a sibling), never skills/<skill>/...: the skills are installed as a plugin, so the user's working directory is their project, not this repo
  • Before opening a PR: claude plugin validate . and uv run --with pytest --with requests --with python-dotenv pytest tests/
  • Authentication: HUBSPOT_ACCESS_TOKEN loaded from .env via python-dotenv. The token is a service key; legacy private app tokens still work but can no longer be created (since October 26, 2026), so never tell a user to create a private app
  • CSV audit logs go to data/audit-logs/ (gitignored — may contain PII, never commit)
  • API targets: existing scripts use /crm/v3/, /automation/v4/, /marketing/v3/. v4 support ends March 30, 2027; v1-v3 are unsupported since September 2026 with enforcement in September 2027. New scripts target the current date-based version (2026-09); existing scripts migrate one skill at a time, each verified with /sandbox-self-test
  • Rate limits: 100 requests per 10 seconds per app on Free/Starter, 190 on Professional/Enterprise; the Search API is capped separately at 5 requests per second
  • See CONTRIBUTING.md for the full skill template, frontmatter spec, and safety-mechanism requirements

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.

Reviews can't be read right now.

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.

Check your setup, free