agentsclimarketplace
Sign in

Secretbroker

Skill HamedMP/secretbroker/skills/secretbroker

Keeps API keys, tokens, passwords, credential paths, and other secrets out of AI conversations while granting explicitly named environment variables to local child commands. Use when a local CLI command needs sensitive credentials or when a user is about to paste a secret into chat.From its SKILL.md

Install
npx -y skills add HamedMP/secretbroker --skill secretbroker

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

  • 26 days oldThe repository was created 26 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 file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

1.6 KB, 256 tokens by cl100k_base, as published. Nobody here has run it

SecretBroker

SecretBroker gives a child command access to explicitly named credentials while keeping values outside the conversation, command arguments, and ordinary tool output.

Mandatory rules

  1. Never ask the user to paste, type, or reveal a secret value in chat.
  2. Never put a secret value in a command, command argument, generated file, message, log, or tool result.
  3. Request all known missing variable names together.
  4. Pass only variable names and non-sensitive descriptions to SecretBroker.
  5. Use secretbroker run --with NAME to run commands that need stored variables.
  6. Never execute env, printenv, shell tracing, credential-display commands, or equivalent inspection under SecretBroker.
  7. Never claim that output redaction or SecretBroker prevents a granted executable from deliberately exfiltrating credentials.

Workflow

Read references/workflow.md, then follow its local executable selection and request workflow.

Before using credentials, read references/security.md and apply its command-boundary rules.

Expected response behavior

Tell the user which variable names are needed and why, never their values. After the request completes, report only readiness by name and continue with the intended command through SecretBroker.

What ships with it: 6 files

51.6 KB alongside SKILL.md

agents/

assets/

evals/

references/

Keep looking

Skills are one crate of 326,144. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.