snitch

What the scan recorded

Observed events

every prompt, subagent spawns, ? calls, classic hooks, every tool call

Hooks

session.start, command.run {command=clear|resume}, turn.step, prompt.submit, turn.start, agent.spawn, turn.complete, session.measure, prompt.compose, tool.call {tool=?}, tool.call {tool=?}, classic.Stop, ui.render {component=ToolUse, props has {tool=?}, ui.render {component=ToolResult, props has {tool=?}, tool.call, command.run {command=?|?}, ui.render {component=PromptHint}, ui.render {component=AbovePrompt}, ui.close {id=?}, ui.render {component=Pane}

API calls

$.agent.list, $.clock.every, $.clock.now, $.command.register, $.fs.read, $.model.complete, $.process.run, $.prompt.submit, $.session.id, $.store.delete, $.store.get, $.store.keys, $.store.set, $.tool.register, $.ui.close, $.ui.copy, $.ui.focus, $.ui.invalidate, $.ui.open, $.ui.panes, $.ui.resolve, $.ui.scroll, $.ui.status, $.ui.toast

Validation on 2.1.304

Passed

Scanned revision

ccab5de1eaf9

Marketplace passed

snitch

UI rewrite review

UI hook source contains control-character strings. Review any values passed to next() as rewritten props.text; the scanner does not trace whether these strings reach that call. hooks/work-files/git.ts:64

These are the hooks and API calls reported by claude plugin validate. Access levels describe the widest reach of those calls and are not safety ratings. Validation is a static check, not a runtime test, and a passing result does not prove that a mod works or is safe. What a mod can access and how the scan works.