Bypasses / Approval or confirmation prompt
Approval or confirmation prompt
36 cases that got past this protection.
- PraisonAI: Human-in-the-loop tool approval is cached by tool name and silently reused for all subsequent calls with arbitrary arguments Approval granted once for a tool name is cached and reused for every later call with any arguments, so later file writes skip review.
- PraisonAI: SkillTools Executes Scripts Without Path Containment Validation The script-running tool executes any file path the model supplies with no containment check, and its approval requirement can be disabled through configuration.
- Pydantic AI Web chat UI (`Agent.to_web()`, `clai web`): a website visited by the developer can trigger agent runs and server-side tool execution on the local chat endpoint A malicious webpage sends cross-origin requests to the local chat endpoint, which runs the agent and its tools without checking content type or approval.
- n8n: Anonymous Approval-Gate Bypass via Reused resumeToken over the Chat WebSocket A reused resume token from a public form lets an anonymous party resume a paused workflow through the chat route, bypassing the approval gate that should hold it.
- Hugging Face Transformers downloads custom generation code before trust consent Remote custom generation code is written into the local module cache before the trust prompt is shown, so declined code still persists on disk.
- praisonaiagents: ast_grep_rewrite rewrites arbitrary files without the @require_approval gate enforced on every sibling mutation tool An AI-callable rewrite tool omits the approval decorator used by sibling file-changing tools, so the model can rewrite files without confirmation.
- PraisonAI: Unauthenticated RCE via Jobs API + Approval Bypass An unauthenticated job endpoint accepts workflow YAML whose top-level approve field disables the approval gate, letting the agent run dangerous shell tools.
- PraisonAI: HTTPApproval dashboard renders tool arguments as raw HTML, allowing approval-page XSS to approve dangerous tools Attacker-influenced tool arguments are inserted unescaped into the HTTP approval dashboard, so a script there can approve a dangerous pending tool call.
- PraisonAI recipe workflow policy can be bypassed by declaring and YAML-approving dangerous tools outside TEMPLATE.yaml A recipe can declare a dangerous shell tool in its workflow file and self-approve it, skipping the policy check so untrusted commands run.
- PraisonAI DiscordApproval accepts unrelated channel messages as dangerous-tool approvals Any non-bot message in the approval channel reading yes counts as approval for a pending dangerous tool call, since replies and approvers are not bound.
- Open WebUI: Cross-origin postMessage confirmation bypass via action:submit The chat page message listener accepts prompt and submit messages from any origin, so an external page can submit prompts and trigger model requests without the confirmation dialog.
- Claude Code: Out-of-Band Data Exfiltration via Pre-Approved HuggingFace Domain in WebFetch Injected content can make Claude Code fetch attacker-chosen Hugging Face URLs without a prompt, leaking data through the request traffic as a covert channel.
- Diffusers: TOCTOU Trust Remote Code Bypass A race between the trust check and the load step lets a Hub repository's custom pipeline code run without the trust_remote_code flag.
- GitHub Copilot CLI: Nested Bare Repository Can Execute Arbitrary Commands via core.fsmonitor A nested bare git repository in a project sets executable config keys, so when the agent runs ordinary git commands it executes attacker commands without asking the user.
- Claude Code: Trust Dialog Bypass via Git Worktree Spoofing Allows Arbitrary Code Execution A crafted repository's worktree commondir file points to an already trusted path, so the trust dialog is skipped and repo-defined hooks run immediately.
- OpenAI Codex CLI enables code execution through malicious MCP (Model Context Protocol) configuration files Running the Codex CLI in a malicious repository auto-loads project config that defines commands, which execute without user confirmation.
- PraisonAI: Unauthenticated Allow-List Manipulation Bypasses Agent Tool Approval Safety Controls An unauthenticated caller can add dangerous tools to the approval allowlist when no token is set, so the agent runs them without human confirmation.
- PraisonAI: Hardcoded `approval_mode="auto"` in Chainlit UI Overrides Administrator Configuration, Enabling Unapproved Shell Command Execution A UI module forces auto-approval over the administrator's manual setting, so the agent runs shell commands without the required confirmation.
- PraisonAI: Coarse-Grained Tool Approval Cache Bypasses Per-Invocation Consent for Shell Commands Approval is cached per tool name rather than per arguments, so later shell commands run without consent and can leak environment secrets.
- PraisonAIAgents: Environment Variable Secret Exfiltration via os.path.expandvars() Bypassing shell=False in Shell Tool The shell tool expands environment variables inside arguments even though it avoids a shell, so reviewers see literal variable names while secret values are read and sent out.
- FastMCP: Missing Consent Verification in OAuth Proxy Callback Facilitates Confused Deputy Vulnerabilities The FastMCP OAuth proxy callback does not verify fresh user consent, so a lured browser session yields an authorization code for the attacker's client.
- Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File A malicious repository's committed settings select bypass-permissions mode before the trust dialog appears, so Claude Code runs tools without the user's consent.
- ONNX Untrusted Model Repository Warnings Suppressed by silent=True in onnx.hub.load() — Silent Supply-Chain Attack Silent mode on the hub loader skips the untrusted-repository prompt, so models from any repo load unnoticed and the checksum comes from the same attacker-controlled repo.
- GitHub Copilot CLI Dangerous Shell Expansion Patterns Enable Arbitrary Code Execution Bash parameter expansion operators hide command substitutions inside arguments the safety check treats as read-only, so the agent runs them without approval.
- Claude Code Vulnerable to Command Injection via Directory Change Bypasses Write Protection Untrusted content in context could make Claude Code use a cd command to enter protected directories and write files without the user being asked to confirm.
- Claude Code has a Command Injection in find Command Bypasses User Approval Prompt Untrusted text placed in the context window could steer Claude Code's find invocation past the user approval prompt, so attacker commands run.
- Claude Code Leaks Data via Malicious Environment Configuration Before Trust Confirmation A repository settings file can point the API base URL at an attacker endpoint, and Claude Code sends requests carrying the user's key before the trust prompt appears.
- Claude Code Command Validation Bypass Allows Arbitrary Code Execution Shell parsing errors around $IFS and short flags let commands pass read-only checks, so injected context content can run arbitrary commands without approval.
- Claude Code vulnerable to command execution prior to startup trust dialog A malicious Yarn configuration in a repository executes code when Claude Code runs yarn --version, before the user accepts the directory trust prompt.
- Claude Code can execute commands prior to the startup trust dialog Project code ran at startup before the user accepted the trust dialog when Claude Code was opened in an untrusted directory.
- Claude Code Vulnerable to Arbitrary Code Execution via Plugin Autoloading with Specific Yarn Versions Claude Code runs yarn --version in untrusted directories, and malicious Yarn config can execute code before the trust dialog is shown.
- Claude Code vulnerable to arbitrary code execution caused by maliciously configured git email A malicious git user.email value in a repository's config is interpolated into a startup shell command, running it before the user accepts the workspace trust dialog.
- Claude Code rg vulnerability does not protect against approval prompt bypass Attacker-supplied text placed into the agent's context exploits a command-parsing flaw so an untrusted shell command runs without the confirmation prompt.
- Claude Code Vulnerable to Arbitrary Code Execution Due to Insufficient Startup Warning Accepting the folder trust prompt silently allowed the agent to run files in that folder, because the warning did not disclose this behaviour.
- Claude Code's Permissive Default Allowlist Enables Unauthorized File Read and Network Exfiltration in Claude Code An overly broad safe-command list let injected content read a local file and send it over the network without confirmation.
- Claude Code echo command allowed bypass of user approval prompt for command execution A command parsing error let an echo invocation skip the confirmation prompt, so injected untrusted content could run arbitrary commands.