AGENT SKILL

guardrails

How AgentGuards enforces its guardrails in Codex — fully automatic, via hooks, with nothing for you to call. Load this if asked how AgentGuards works, whether you need to screen something yourself, or why no agentguards check_input / authorize_action tools are available.

License 许可
Apache-2.0
最近更新
2026-11-14
Stars
★ 1.6k
Forks
491

01 前言 Front Matter

name
guardrails
description
How AgentGuards enforces its guardrails in Codex — fully automatic, via hooks, with nothing for you to call. Load this if asked how AgentGuards works, whether you need to screen something yourself, or why no agentguards check_input / authorize_action tools are available.

03 SKILL.md 正文

AgentGuards — enforced by hooks

This plugin has no MCP server, so there is no check_input, authorize_action or health_check tool — do not look for them, they are not installed.

That is not reduced protection. The hooks in hooks/hooks.json enforce deterministically, on every request, whatever the model does:

  • UserPromptSubmit screens every prompt before you see it.
  • PreToolUse authorizes every shell command before it runs: denied, or allowed, or left to Codex's own approval prompt.
  • PermissionRequest rides that approval prompt: it denies what AgentGuards rejects and remembers what the user approved for the rest of the session.
  • PostToolUse scans the output of shell commands that fetch (curl, wget, http, fetch, aria2c — also behind sudo, timeout, bash -c, $(...)), withholding or redacting it before you read it, and scans apply_patch edits for vulnerabilities and secrets.

Nothing here requires you to call a tool, check a decision, or format a block message — the hook has already acted by the time you see (or don't see) the result. If a hook blocks something, reply with its message rather than composing your own.

One limit, stated plainly: Codex's built-in web search runs on OpenAI's servers, not on this machine, so no hook ever sees its results. Content fetched with a shell command is screened; built-in web search results are not.

To check that the guardrails are on, use the status skill. To set the API key, use the setup skill.

04 同仓库更多 Skills

hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-review-with-annotations

Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-find-and-synthesize-context

Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-create-or-update-docs

Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-create-or-manage-records

Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-create-or-manage-interactive-html

Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-collaborate-threads

Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.

★ 1.6k