Workflow skills for AI agents with APort passports. Drop these into your agent's skills directory — Claude Code, OpenClaw, Cursor, or any framework that reads a skills folder.
These are not prompts. They are skills that interact with real APort infrastructure — verified identity, enforced deliverable contracts, signed receipts.
Open your AI agent and paste this:
Install aport-skills: run
git clone https://github.com/aporthq/aport-skills.git ~/.claude/skills/aport-skills && cd ~/.claude/skills/aport-skills && ./setup
Or for a specific project so teammates get it:
Add aport-skills to this project: run
cp -Rf ~/.claude/skills/aport-skills .claude/skills/aport-skills && rm -rf .claude/skills/aport-skills/.git
| Skill | Command | What it does |
|---|---|---|
| APort ID | /aport-id |
Register yourself with APort — get a verifiable passport (DID credential) with identity, capabilities, and deliverable contract |
| APort Complete | /aport-complete |
Verify a completed task against your deliverable contract before marking it done. Enforces quality gates deterministically. |
| APort Standup | /aport-standup |
Generate a standup update from your signed APort receipts. What you actually shipped, not what you remember. |
| APort Handoff | /aport-handoff |
Package completed work for handoff to another agent or human. Every item backed by a verified receipt. |
| APort Status | /aport-status |
Show your passport, capabilities, deliverable contract, and recent activity. Like git status for your accountability state. |
You start a session → /aport-status (what am I, what can I do)
You do work → [your normal workflow]
You finish a task → /aport-complete (verify before marking done)
You sync with the team → /aport-standup (what did I actually ship)
You hand off work → /aport-handoff (verified work package)
A new agent joins → /aport-id (get your passport)
Skills like gstack tell an agent what cognitive mode to use. These skills enforce what an agent must deliver before it can call itself done.
gstack's /review says: "think like a paranoid staff engineer."
APort's /aport-complete says: "you cannot mark this done until APort verifies
you met your deliverable contract."
The prompt changes the behaviour. The policy enforces the outcome.
Both are useful. They complement each other.
You need an APort passport to use most of these skills.
Web: https://aport.id
CLI: npx aport-id
Agent: Send your agent to aport.id/skill
When you claim your passport (via the email link), an API key is automatically generated and shown on the confirmation page. Save it immediately — it's shown only once.
The key lets your agent read its own passport and verify tasks:
GET https://aport.io/api/passports/YOUR_AGENT_ID
POST https://aport.io/api/verify/policy/deliverable.task.complete.v1
Authorization: Bearer YOUR_API_KEY
The key has read and status scopes. Agents cannot update their own passport — that's the owner's job. To update a passport, visit https://aport.id/manage or log in at https://aport.io/dashboard with the email used to claim.
Store the key in aport-passport.json (already in .gitignore) or your environment as APORT_API_KEY.
- aport.id — https://aport.id
- APort platform — https://aport.io
- Policy packs — https://github.com/aporthq/aport-policies
- Passport spec — https://github.com/aporthq/aport-spec
- aport.id source — https://github.com/aporthq/aport-id
Apache 2.0