The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The scripts/install.sh script downloads a payload from 'https://plannotator.ai/install.sh' and pipes it directly into the bash interpreter. This is a high-risk pattern that allows a remote server to execute arbitrary code on the host system without prior verification.
[EXTERNAL_DOWNLOADS]: The skill requires downloading external CLI tools and plugins from the plannotator.ai domain, which is not included in the trusted vendors list.
[COMMAND_EXECUTION]: Multiple scripts perform persistent system modifications, including appending environment variables to shell profiles like .zshrc and .bashrc. It also modifies AI agent configuration files in ~/.claude/, ~/.gemini/, and ~/.codex/ to register automated execution hooks.
[PROMPT_INJECTION]: The scripts/setup-codex-hook.sh script injects new procedural instructions into the 'developer_instructions' field of the Codex configuration. This forces the agent to follow a specific plan-review workflow and interact with the plannotator tool for every task.
[DATA_EXFILTRATION]: The skill facilitates the transfer of agent-generated plans and code diffs to external applications like Obsidian and Bear. It also supports a custom share portal URL (share.plannotator.ai) which could potentially be used to upload sensitive project plans to an external server.

plannotator