Install
$ agentstack add skill-zhjai-agent-completion-gate-completion-gate-init ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Completion Gate Init
Convenience wrapper around the authoritative setup script scripts/init.sh. Use it to scaffold the gate into the user's project. The script is the source of truth — run it, don't reimplement it.
When to use
- The user asks to set up / initialize / add the completion gate to their repository.
Procedure
- Locate an
agent-completion-gatecheckout (the engine + scaffolder live in the repo, not
in this installed skill — npx skills add ships skills, not the gate engine). If the user doesn't have one, clone it: `` git clone https://github.com/zhjai/agent-completion-gate ``
- From the user's project root, run the scaffolder, pointing it at the project:
`` sh /scripts/init.sh --dest . ` It creates gate/ (engine + empty manifest), control/surface_inventory.yaml, state/, .github/workflows/completion-gate.yml, and a CODEOWNERS example. It is idempotent and will not overwrite existing files (engine or specs) without --force`.
- Relay the script's "Next" steps to the user verbatim. Then help with step 1 of those (define
real surfaces/checks) if asked.
Do not
- Do not edit
gate/acceptance_manifest.yamlto weaken it, do not movegate//control/
into a worker-writable location, and do not invent your own scaffolding — only run the script.
- Do not claim the gate is active/authoritative. It only enforces the state machine until the
user (a human) sets the required status check + CODEOWNERS. Say so plainly.
- Do not set up anything that lets you (the agent) grant
complete. That asymmetry is the
whole point: you propose candidate_complete; CI / a human verifies.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: zhjai
- Source: zhjai/agent-completion-gate
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.