Deploy app to agents

Skill saezbaldo/deploytoagents-skills/deploy-app-to-agents

Portable Agent Skills for verifiable app distribution and LLM discovery

Install

npx -y skills add saezbaldo/deploytoagents-skills --skill deploy-app-to-agents

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

16 days oldThe repository was created 16 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Audit and distribute an existing app, API, SaaS, MCP server, or agent so AI assistants can find, verify, recommend, and invoke it. Use when a user asks how to launch or distribute software to agents or LLMs, improve agent discoverability, publish MCP/A2A metadata, create machine-readable discovery surfaces, compare distribution channels, or measure whether unbranded prompts mention their product.

SKILL.md

2.7 KB, 429 tokens by cl100k_base, as published. Nobody here has run it

Deploy an app to agents

Use Deploy to Agents as the control plane for a verifiable distribution workflow. Do not promise model inclusion, rankings, or favorable recommendations.

Workflow

Identify the product URL and the real user problems for which it should be found.
Audit the public URL with the Deploy to Agents MCP server at https://deploytoagents.com/mcp or run npx deploytoagents audit <url> --json.
Separate findings into technical surfaces the agent can implement, registry submissions requiring owner approval, and independent evidence that must be earned rather than generated.
Create the distribution plan through MCP or the authenticated CLI. Preserve accurate limitations and authentication requirements.
Implement or prepare the recommended artifacts. Never publish externally, create accounts, or make endorsements without the owner authorizing that action.
Verify every published artifact from its public URL or external registry record.
Run fresh, unbranded discovery prompts. Record successes, failures, cited sources, model labels, and dates in Discovery Lab.
Recommend iteration based on observed discovery evidence, not metadata completion alone.

Evidence rules

Classify the product's own pages as owned evidence.
Classify official registries, package registries, signed releases, and customer-controlled integration pages as externally verifiable evidence.
Treat editorial coverage, substantive community discussion, independent benchmarks, and unbranded recommendations as independent evidence only when the publisher is genuinely independent.
Disclose ownership, commercial relationships, sponsorship, or incentives.
Retain negative discovery results. Do not manufacture testimonials, votes, discussions, backlinks, or model answers.

Agent handoff

Return structured output containing:

the product and capability being distributed;
current channels and evidence URLs;
ordered automatic, agent-executable, and manual steps;
expected artifact for each step;
a public verification method;
unresolved risks or claims that lack external proof.

Use npx deploytoagents login before accessing a customer's private portfolio or recording Discovery Lab observations. Public audits and public product evidence may be inspected without customer credentials.

What ships with it: 1 file

482 B alongside SKILL.md

agents/

openai.yaml482 B