Toolkit
Lifecycle Hooks · Pitfalls
On this page7 sections
Pitfalls
Claiming Enforcement from Event Coverage
A product may fire a hook without granting a blocking decision. Verify the output contract and failure behavior.
Ignoring Timeouts
A handler that fails open on timeout is an availability guardrail, not a hard security boundary. Keep it fast and move final authorization elsewhere.
Missing Alternate Paths
Blocking one shell command does not block the equivalent file tool, MCPUSB-C for AI — one standard plug so any agent can connect to any data source or tool.The Model Context Protocol: an open standard (Anthropic, 2024) for connecting AI systems to external tools and data without building a custom integration for each one.Full definition action, nested AgentA worker you delegate to: brief it, and it takes steps on its own. A chatbot answers; an agent acts.An LLM that runs tools in a loop toward a goal — it acts, checks the result, decides the next step, and repeats until done.Full definition, IDE operation, or cloud executor.
Recursive Hooks
A completion hook that triggers more agent work can trigger itself again. Use explicit re-entry markers and bounded retries.
Untrusted Repository Scripts
Hook configuration and scripts checked into an untrusted repository can execute with developer credentials. Require review and managed allowlists before activation.
Non-Deterministic Decisions in a Deterministic Slot
PromptThe brief. Closer to writing instructions for a new contractor than to programming.The text you give the model — question, instructions, context. Its quality shapes the quality of the output.Full definition or agent handlers can add judgment, but they do not turn the hook into deterministic policy. Label model-derived decisions and retain the evidence.
Falsification Checklist
- Does the handler still control the action when it crashes or times out?
- Do nested agents and MCP tools traverse the same event?
- Can repository code disable or replace the hook?
- Is the release check repeated outside the local agent runtime?
Source register
Lifecycle Hooks sources
Primary vendor documentation reviewed on the date shown. Links may change after publication.
- Hooks reference (opens in a new tab)
Anthropic · official docs · accessed 2026-08-01
- Hooks (opens in a new tab)
OpenAI · official docs · accessed 2026-08-01
- GitHub Copilot hooks reference (opens in a new tab)
GitHub · official docs · accessed 2026-08-01