agentsclimarketplace
Sign in

Marineconnect qc reporting

Skill Marineconnect/marineconnect-agent-skills/plugins/marineconnect-ops/skills/marineconnect-qc-reporting

Create MarineConnect MCP QC evidence reports. Use when the user asks to QC tools, show prompt/tool/params/output, verify end-to-end agent behavior, run read-only smoke checks, compare expected vs actual tool routing, produce checklist evidence, or avoid blind QC for MarineConnect MCP and operations skills.From its SKILL.md

Install
npx -y skills add Marineconnect/marineconnect-agent-skills --skill marineconnect-qc-reporting

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

3 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.
  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 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.

SKILL.md

1.4 KB, 228 tokens by cl100k_base, as published. Nobody here has run it

MarineConnect QC Reporting

QC Standard

Every QC result must show prompt or question, selected tool, params, result evidence, assertion, and pass/fail.

Workflow

  1. Start with protocol checks when MCP connectivity is in scope: initialize, tools/list, or mcp_health_check.
  2. For agent behavior, run prompt -> mcp_list_tools_by_role discovery -> resolver -> downstream tool chain -> assertions.
  3. Use only MCP read tools for standalone plugin QC. Do not assume that a private backend checkout or direct database access is available.
  4. Authorized maintainers may use backend-owned QC scripts only when a separate backend checkout provides them and that repository's safety rules are in force.
  5. Never request database credentials, bypass OAuth, or turn public plugin QC into a direct production database session.
  6. Write or reference reports under docs/ when a writable project is in scope, and summarize failures first.

Output

  • Summary counts, failed/skipped tools, E2E prompt chains, full tool table link, and explicit caveats for zero-row analytics results.

What ships with it: 1 file

495 B alongside SKILL.md

agents/

Keep looking

Skills are one crate of 326,422. 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.