skills.rest
Zhachory1
Zhachory1COMMUNITY
Explore all skills by this author →
Website𝕏@zhachory1GitHub

human-approval-gate

Prepares decision packages with evidence, risks, and rollback plans for human signoff.

Updated Jul 8, 2026
Search Tags:#risk assessment#escalation#human approval#rollback plan#signoff#workflow gate#decision package
One-click install
npx skills add https://github.com/Zhachory1/autopraxis --skill human-approval-gate-zhachory1
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: human-approval-gate
Source: https://github.com/Zhachory1/autopraxis/tree/main/skills/human-approval-gate
Command: npx skills add https://github.com/Zhachory1/autopraxis --skill human-approval-gate-zhachory1

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? High-stakes decisions like merges, deploys, roadmap commitments, and destructive actions need accountable human ownership, but agents often bury the ask in prose or auto-approve. This Skill structures a clear yes/no decision package so a qualified human can decide quickly with full context. ## Core Features & Use Cases - Structured Decision Package: Produces a concise ask with recommendation, confidence, evidence, risks, alternatives, and rollback plan. - Escalation on Budget Caps: Stops agent loops when budget or evidence limits are hit and escalates to a human instead of churning. - Dissent Visibility: Keeps council dissents and unresolved concerns visible rather than hidden in summaries. - Use Case: After a PR review workflow completes, use this Skill to present a merge/no-merge decision to the engineering lead with test results, review verdicts, and rollback options. ## Quick Start Ask the agent to prepare a human approval request for the pending production deployment including evidence, risks, and a rollback plan.

Frequently Asked Questions about human-approval-gate

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get human signoff before merging a PR in an agent workflow?▼

Use a human approval gate to convert the review result into one explicit merge/no-merge ask. The package includes the recommendation, confidence, evidence links, risks, and rollback plan so the accountable owner can decide directly.

What should a deployment approval request include?▼

A deployment approval request should include one clear decision ask, the decision owner, a recommendation with confidence, supporting evidence such as tests and metrics, risks with mitigations, alternatives, and a rollback plan.

When should an AI agent escalate to a human instead of continuing?▼

Escalate when budget or loop caps are reached, evidence is exhausted, or the action involves production risk, destructive changes, or irreversible commitments. The agent may recommend but never auto-approves accountable decisions.

Can an AI agent approve its own deployment or merge decisions?▼

No. The agent may only recommend with stated confidence and conditions; a qualified human owns the final decision. This separation preserves accountability for high-risk, irreversible, or budget-bound actions.

Why do approval requests get ignored or delayed by reviewers?▼

Approval requests fail when the ask is buried in prose, evidence is missing, or certainty is overstated. Putting the decision first, linking proof, and stating confidence and caveats makes requests actionable.

Related Skills

Explore More Product & Management Skills→
maxwell2732maxwell2732
39

research-harness-release

Assemble and validate a replication package after an APPROVE verdict.

Community
Advanced
me2reshme2resh
489

approve-merge

Record CEO approval markers and merge GitHub PRs with squash.

Community
Advanced
nexus-labs-automationnexus-labs-automation
7

human-in-the-loop

Capture human approval, clarification, and escalation events as structured spans in AI agent workflows.

Official
Advanced
razorpayrazorpay
31

risk-assessment

Classify pull requests in the agent-skills repository as LOW, MEDIUM, or HIGH risk.

Official
Intermediate