Codacy launches agent skills and cloud CLI for AI coding assistants
Codacy released agent skills and a cloud CLI that let AI coding assistants like Claude handle code quality and security tasks — such as fixing failing quality gates, adding tests, and suppressing false positives — without consuming agent tokens.
Score breakdown
The release allows AI coding agents to autonomously manage code quality gate workflows server-side, removing the need for manual UI interaction and avoiding agent token consumption.
- 01Codacy is a code quality and security platform for enforcing coding standards on AI-generated code.
- 02The launch includes agent skills and a cloud CLI compatible with assistants like Claude, Codex, and Copilot.
- 03Agents can fix failing quality gates, add tests, and suppress false positives with a reason via natural-language prompts.
Codacy, a code quality and security platform focused on helping engineering teams enforce coding standards against AI-generated code, has launched agent skills alongside a cloud CLI. The release enables AI coding assistants — including Claude, Codex, and Copilot — to handle routine code quality and security tasks through natural-language prompts rather than manual UI interaction.
A representative example prompt given in the announcement is: "PR 42 is failing the gate, fix what's real, add the tests, ignore the false positives with a reason, re-run." This illustrates how an agent can orchestrate multiple quality-gate actions in a single instruction. A notable technical detail is that analysis runs server-side, meaning the workflow does not consume agent tokens.
Key facts
- 01Codacy is a code quality and security platform for enforcing coding standards on AI-generated code.
- 02The launch includes agent skills and a cloud CLI compatible with assistants like Claude, Codex, and Copilot.
- 03Agents can fix failing quality gates, add tests, and suppress false positives with a reason via natural-language prompts.
- 04Tasks are handled without requiring use of the Codacy UI.
- 05Analysis runs server-side and does not consume agent tokens.
Topics
Summary and scoring are generated automatically from the original article. We always link back to the publisher and never republish images or paywalled content. Last processed Jun 10, 2026 · 15:34 UTC. How this works →