Skills

A skill is a set of instructions an agent reads before it acts. Deliverd's 10 are open source under the MIT licence, in deliverd-dev/deliverd-skills on GitHub. They teach judgement — when to ask, how to write a request, what to do with the answer — and use the MCP server, the SDKs or the CLI when one is set up. Each also works in plain conversation when none is.

The skills#

Approval

When to stop and ask a person before a consequential action — anything irreversible, anything that spends money or reaches people outside the conversation — how to write a request someone can decide from a phone, how to wait, and how to act on the answer, including an approver's corrections.

Ethics

Noticing when work could harm or wrong a person, and putting that concern in front of whoever decides rather than around them. It also covers behaving well when an organisation's ethics rules flag or refuse a request, and helping an administrator write those rules.

Ethics rules

Helping a person write their organisation's ethics rules: finding where rules belong, adapting the starter rules and a library grouped by area, keywords that don't catch everyday software talk, testing them, and switching them on one at a time. The agent drafts; a person adds them.

Artifact

Building a well-made, self-contained HTML page — a report, a dashboard, a plan, a prototype — and then offering to publish it. The craft half works with no account at all.

Tasks

Keeping long work aimed at its goal: stating an objective, keeping a short working state instead of relying on the history, branching a problem into a side task and bringing its result back, and leaving decisions to a person.

Compliance

Doing AI work under an organisation's governance and leaving a record an auditor can check: the governance profile, controls met by evidence and scored evaluation results, the release the evidence tested, the release check before anything consequential, incidents and DORA reports, and what only a person may decide. It never calls the work compliant.

Screen

Checking text for personal data, secrets and confidential terms before it goes to a model or leaves the machine, with Deliverd Screen running locally: wiring the check into code that calls a model, redacting what is handed on, and acting on the answer — send, send redacted, ask a person, or send nothing. Free, with no account.

Prompts

Using the organisation's approved prompts instead of its own words when one fits the job: finding one, taking the version in force filled in with the right inputs, checking the result against its limitations, and drafting words that worked as a new prompt for a person to approve.

Policy

Helping a person write the rules that decide what agents may do on their own, what goes to a person and what is refused: proposing a setup, naming the actions, policy files with test cases, replaying a change against past decisions, and the CI that checks every change. The agent drafts rules; a person puts them in force.

Checkpoint

Putting the approval step on the path to a consequential action instead of relying on the agent to ask: the hosted MCP gateway, a gate in the code that acts, checking the signed decision token where the action happens, the GitHub Action and the AgentCore Gateway interceptor, and saying plainly what each one does not cover.

Install one#

Claude Code, Codex and Cursor read skills from the same place, so one install covers all three. Install the ones you want:

npx skills add deliverd-dev/deliverd-skills --skill approval

npx skills add deliverd-dev/deliverd-skills --skill ethics

npx skills add deliverd-dev/deliverd-skills --skill ethics-rules

npx skills add deliverd-dev/deliverd-skills --skill tasks

npx skills add deliverd-dev/deliverd-skills --skill compliance

npx skills add deliverd-dev/deliverd-skills --skill screen

npx skills add deliverd-dev/deliverd-skills --skill prompts

npx skills add deliverd-dev/deliverd-skills --skill policy

npx skills add deliverd-dev/deliverd-skills --skill checkpoint

npx skills add deliverd-dev/deliverd-skills --skill artifact

Approval and ethics are written to work together: the ethics skill decides that a person should weigh something, and the approval skill is how the agent asks.

The Claude Code plugin#

In Claude Code, the plugin installs the Deliverd MCP server and all 10 skills at once. Add the marketplace, then install the deliverd plugin it lists from /plugin:

claude plugin marketplace add deliverd-dev/deliverd-skills

The first time a tool runs, Claude Code opens a browser for you to approve the connection. See Authentication.

What they hold to#

• Everything the approval skill names — MCP tools, request and response fields, CLI flags, SDK environment variables — is tested against the code before each release, so it teaches what the product does.
• The ethics skill's principles, effects and refusal codes are tested the same way against the ethics module. See Ethics concerns.
• The ethics rules skill's library is run through the same validator and matcher the ethics module uses: every rule must be accepted, catch its examples and leave its counter-examples alone.
• The artifact skill's examples are run through the same scanners a publish runs, so it does not teach a page the platform would refuse.
• Every tool, field, SDK method and CLI command the tasks skill names is tested against the code the same way. See Objectives.
• The compliance skill is held the same way: every tool, profile and evidence field, report format, gate answer and CLI flag it names, and the list of decisions only a person may make.
• The policy skill's example policy file is run through the same parsing, validation and evaluation deliverd policy test uses, and its own test cases must pass. See Policy as code.
• The prompts and checkpoint skills are held the same way: every tool, field, SDK method, CLI command, gateway address and integration setting they name.

The skills are MIT licensed: read them, fork them, or lift the parts you want. The service they optionally publish to is not open source.