Agentic AI Atlasby a5c.ai
OverviewWikiGraphFor AgentsEdgesSearchWorkspace
/
GitHubDocsDiscord
iiRecord
Agentic AI Atlas · Claude Code stop hook can block (deny) at end-of-turn
claim:stop-hook-claude-interactive-blocksa5c.ai
Search record views/
Record · tabs

Available views

II.Record viewspp. 1 - 1
overviewjsongraph
II.
TestableClaim overview

claim:stop-hook-claude-interactive-blocks

Reference · live

Claude Code stop hook can block (deny) at end-of-turn overview

A hooks-mux handler returning decision: deny at turn.stop propagates the deny decision in the output for Claude Code, since the Claude adapter supports the blockCapability. This allows hooks to prevent the agent from stopping (e.g., to enforce a review step before turn completion).

TestableClaimOutgoing · 3Incoming · 2

Attributes

displayName
Claude Code stop hook can block (deny) at end-of-turn
description
A hooks-mux handler returning decision: deny at turn.stop propagates the deny decision in the output for Claude Code, since the Claude adapter supports the blockCapability. This allows hooks to prevent the agent from stopping (e.g., to enforce a review step before turn completion).
testStatus
pending
testType
integration
testCommand
npx vitest run packages/atlas/tests/claims/stop-hook-control.test.ts -t "claude adapter.*deny decision propagates"
category
hooks
priority
high

Outgoing edges

asserts_about2
  • agent:claude-code·AgentProductClaude Code
  • hook-surface:stop·HookSurfaceStop (canonical)
tests_hook_surface1
  • hook-surface:claude.stop·HookSurfaceStop (Claude)

Incoming edges

backed_by_evidence1
  • evidence:hooks-mux-stop-hook-test-suite·EvidenceSourcehooks-mux stop-hook control test suite
tests_claim1
  • experiment:stop-hook-control-interactive·ExperimentStop Hook Control — Interactive Mode

Related pages

No related wiki pages for this record.

Shortcuts

Open in graph
Browse node kind