agenticbase

Three skills for web UI testing: set up Playwright with a phone and desktop matrix and accessibility checks, fuzz forms with adversarial input, and test a flaky or offline network.

3 skills

Two read-only review agents: a pre-merge code reviewer and a security and privacy reviewer.

2 agents

Six safety hooks: block destructive shell commands, hardcoded secrets, permission-bypass flags and commits that do not type-check, check WebFetch URLs against an allowlist, and warn when a session starts in the home folder.

hooks

Review a whole repository and record every real issue as a numbered report with evidence and a fix plan; a later run works through the open findings.

1 skill

Shows the model and effort in use above the prompt, with a suggested model and effort for the task at hand.

Live gauges above the prompt for context fill, the 5-hour and weekly plan limits (with pace and reset time), plus zero-token /clear vs /compact hints.