No Shell Agent Operating Pack — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited No Shell Agent Operating Pack (Agent Skill) and scored it 92/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 2 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 2 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
An open-source operating pack for people who already use AI agents but keep getting "empty-shell automation": files, buttons, or dashboards that look finished but do not actually run a real workflow.
This is not a prompt-pack that promises magic. It is a plain-language operating system for delegating work to ChatGPT, Claude, Codex, Gemini, Hermes, and similar agents with a real objective, inputs, tool choice, permissioned account route, validation, recovery, and live-action boundary.
This repository now also ships a remote MCP/API product: No-Shell Agent Architect MCP.
Current launch mode: free public beta. There is no payment flow, usage gate, paid API call, or private customer data collection in this version. The goal is to observe GitHub stars, issues, workflow feedback, and real usage before deciding whether a paid version should exist later.
It takes a plain-language automation goal and returns:
Local API preview:
npm install
npm run devThen open:
http://localhost:3000Canonical remote MCP endpoint:
https://ai-automation-operating-pack.vercel.app/api/mcpLive demo:
https://ai-automation-operating-pack.vercel.appLaunch kit:
https://ai-automation-operating-pack.vercel.app/launchPublic beta feedback:
https://github.com/loved0543-dotcom/no-shell-agent-operating-pack/issues/1Machine-readable M2M package contract:
https://ai-automation-operating-pack.vercel.app/api/m2m-packageOfficial MCP Registry:
https://registry.modelcontextprotocol.io/v0/servers?search=no-shell-agent-architect-mcpCompatibility URL:
https://no-shell-agent-architect-mcp.vercel.appThe compatibility URL is kept for old shared links only. New docs, client config, and registry metadata should use https://ai-automation-operating-pack.vercel.app.
Client config:
{
"mcpServers": {
"no-shell-agent-architect": {
"url": "https://ai-automation-operating-pack.vercel.app/api/mcp"
}
}
}free/ai_automation_failure_diagnostic_card.md.demo/before_after_email_document_demo.md.delivery/01_customer_intake.md to describe one real workflow.delivery/02_tool_router.md.delivery/03_command_cards.md.delivery/04_result_scorecard.md.delivery/05_recovery_playbook.md.landing/index.htmlpdf/free-diagnostic-card.pdfpdf/before-after-demo.pdfdist/no-shell-agent-operating-pack-starter-v1.zipdist/no-shell-agent-operating-pack-workbench-v1.zipoutreach/validation_tracker.csvoutreach/public_beta_tracker.csvCollect current public beta signals without any account login or paid API:
npm run collect:betaFor the local operator's Obsidian vault, use:
npm run collect:beta:obsidianDogfood the product against its own public beta operation:
npm run dogfood:beta
npm run check:beta-opsThis writes outreach/dogfood_public_beta_ops.md, stages one public feedback post packet in outreach/public_beta_one_channel_launch.md, and verifies the action ledger and permissioned connector v1 contract.
For agent-to-agent or agency packaging, call the M2M package contract:
Invoke-RestMethod -Method Get -Uri https://ai-automation-operating-pack.vercel.app/api/m2m-packageThis returns the MCP/API surfaces, required input schema, guaranteed output blocks, delivery artifacts, paid-readiness gates, and the current beta-signal risk.
Run the package selfcheck:
npm run selfcheck
npm run test
npm run buildExpected result:
PASS package selfcheck
PASS mcp product selfcheckThe selfcheck verifies required files, rendered PDF/PNG artifacts, ZIP contents, landing copy, README/manifest, validation-tracker fields, MCP metadata, tool coverage, and secret-pattern safety.
It also checks the public beta operations loop: signal collector, dogfood output, one-channel staged launch packet, action ledger, and permissioned connector v1 runbook.
After production deploys, verify the live contract:
npm run check:liveThis checks the canonical URL, compatibility URL, MCP tools/list, and the official Registry latest entry.
Launch copy, community drafts, Product Hunt copy, Show HN copy, Reddit/LinkedIn/X drafts, and directory submission text live in docs/LAUNCH_KIT.md.
Public posting is not a bypass. It may be automated only through an approved API, connector, or logged-in browser session with an allowlisted destination, draft/staging mode, an action ledger, and an exact live-action instruction after checking each community's rules.
Deployment, monitoring, compatibility alias recovery, and post-deploy checks are documented in docs/OPERATIONS.md.
This repository is public, open source, and free to use as a public beta. It is early-stage and still needs real external validation. The product is complete enough to show to users, but demand is not proven until public usage signals and the validation trackers have real responses.
No payment connection, storefront publish, customer credential, secret, or private customer data is included in this repository. The product describes permissioned connector architecture, not stored account access. Future paid packaging is intentionally postponed until there is clear usage and feedback.
MIT. See LICENSE.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.