Best for
- A production deploy, restore, migration, bulk refund, mass email, data
- The user is not the approver or the user asked for a named owner to authorize
- The approver is away from a keyboard, which is exactly when a phone call beats
CALLE-AI/awesome-phone-call-agents/skills/deployment-approval-call/SKILL.md
Place one CALL-E phone call to get a spoken, code-verified human approval before an agent or a pipeline does something irreversible, such as a production deploy, a database restore, a bulk refund or a migration. Use when the approver is away from a keyboard and the action cannot be undone.
Decision brief
Use this skill when you are about to do something you cannot undo and a person has to say yes first.
Compatibility matrix
| Platform | Status | Evidence | What to check |
|---|---|---|---|
| Codex | Not declared | No explicit evidence | Portability before use |
| Claude Code | Not declared | No explicit evidence | Portability before use |
| Cursor | Not declared | No explicit evidence | Portability before use |
| Gemini CLI | Not declared | No explicit evidence | Portability before use |
Installation
The source command is displayed only when detected. A safe inspection prompt is always available so your agent can explain every action before execution.
npx skills add https://github.com/CALLE-AI/awesome-phone-call-agents --skill "skills/deployment-approval-call"Inspect the Agent Skill "deployment-approval-call" from https://github.com/CALLE-AI/awesome-phone-call-agents/blob/a34d6b803ee2f1b80446056dcef38911882726e5/skills/deployment-approval-call/SKILL.md at commit a34d6b803ee2f1b80446056dcef38911882726e5. List every install step, command, network request, credential, file read/write, external action, and rollback step. Explain whether it fits my task. Do not install or execute anything until I approve.
Workflow
A production deploy, restore, migration, bulk refund, mass email, data
The action is reversible and cheap. Ask in chat instead of ringing a phone.
1. You write a request file: the change title, a one line summary, the environment, who requested it and the enrolled approvers in ladder order. 2. You run the gate in preview and show the user the exact call script. 3. On the user's go-ahead you run it live. The gate prints a s…
bash cd apps/typescript/phone-approval-gate npm install
Review the “No call, no credentials. Always do this first.” section in the pinned source before continuing.
Permission review
The documentation asks the agent to create, modify, or delete local files.
You write a request file: the change title, a one line summary, theThe documentation asks the agent to run terminal commands or scripts.
npm installThe documentation asks the agent to run terminal commands or scripts.
npm run gate -- preview --request /tmp/approval-request.jsonEvidence record
| Signal | Value | Evidence type | Meaning |
|---|---|---|---|
| Quality score | 93/100 | Computed | Documentation, specificity, maintenance, and trust rules |
| Repository stars | 66 | Source | Repository attention, not individual Skill quality |
| Compatibility | 0 platforms | Source | Declared in the catalog source record |
| Usage guide | automated source guide | Editorial | Generated or reviewed according to the visible evidence level |
Pinned source
Use this skill when you are about to do something you cannot undo and a person has to say yes first.
It does not invent an approval mechanism. It drives the runnable
phone-approval-gate app, which
places one CALL-E call per approver, reads the change out loud, requires the
person to read back a one-time code shown in this conversation and returns an
exit code plus a hash-chained record.
references/safety.md.cd apps/typescript/phone-approval-gate
npm install
# No call, no credentials. Always do this first.
npm run gate -- preview --request /tmp/approval-request.json
# One call per approver, in order. Needs CALLE_API_KEY in the environment.
# --audit is required: every live run appends one approval record.
npm run gate -- request --request /tmp/approval-request.json --live --audit approvals.jsonl --json
The request file shape, the policy fields and the reason codes are documented in
the app README. policy.mode: "dual" requires two approvals from two different
handsets, which is the control to use for a destructive database action.
| Exit code | What you do |
|---|---|
| 0 | Approved. Proceed with the exact action that was described on the call, nothing more. |
| 10 | A person rejected it. Stop. Tell the user who rejected it and do not re-run the gate. |
| 20 | No approval: no answer, voicemail, wrong code, window closed. Stop and report the reason. If the reason is call_state_unknown, a call may still be live: report the call id and do not run the gate again until a person has reconciled it. |
| 30 | The request file is wrong. Fix it and preview again. Do not place a call to find out. |
After a run, tell the user the verdict, the approver id, the reason when there is
one and the audit record hash. Say plainly that nothing was changed when the
verdict was not approved.
references/examples.md: worked requests and the
replies to give the user.references/safety.md: consent, enrolment, masking,
cancellation and the boundaries this skill will not cross.Frequently asked questions
Use this skill when you are about to do something you cannot undo and a person has to say yes first.
The source record exposes this install command: npx skills add https://github.com/CALLE-AI/awesome-phone-call-agents --skill "skills/deployment-approval-call". Inspect the command and pinned source before running it.
Static rules flagged write-files, exec-script in the source; the page lists the matching lines and excerpts.
Alternatives
vasilyu1983/AI-Agents-public
Guides iOS testing with XCTest, XCUITest, Swift Testing, simctl, and xcresult. Use when choosing destinations, controlling flakes, or parsing test artifacts for native apps.
garrytan/gbrain
Generate a publication-quality PDF from any brain page via the gstack make-pdf binary. Strips YAML frontmatter, sanitizes emoji, applies running headers and page numbers. Brain page is always the source of truth; PDF is a rendering.
NVIDIA/skills
How to swap the DeepStream CV detection model in the VSS Alerts Blueprint verification (2d_cv) mode - covers ONNX export, custom bbox parsers, compose mount gotchas, nvinfer config, runtime TRT engine build, deployment, and a segmentation-capable model addendum handoff.
vasilyu1983/AI-Agents-public
Scans public GitHub repos for agent skills, dev practices, and code patterns. Use when enriching skills, setting team policy, or researching a build domain.