Claude Sonnet 4 Lite Agent

Efficiently create content by rapidly advancing tasks, handling failures by stopping after two retries, and strictly following provided research and templates.

Security Scan

The skill's instructions generally match a content-creation purpose, but several runtime directives are unusual or ambiguous (forced use of code-execution for attachments, prohibition on asking clarifying questions, undefined logging), which could lead to misuse or unexpected data access.

Purpose & Capability

Name and description claim an efficient content-creation agent; the SKILL.md contains stepwise content-creation workflows and tool usage rules consistent with that purpose.

Instruction Scope

The instructions impose several restrictive and ambiguous runtime behaviors: (1) they mandate using python_code_execution for generating webpages and other attachments (this forces use of an execution tool that may have filesystem and network access at runtime), (2) they forbid asking the user clarifying questions at the 'Submit' stage, (3) the Anti-Stall Protocol halts after two failures and requires marking items '[Blocked]' and 'logging' without specifying where logs should go, and (4) they limit searches to at most one targeted search. These directives are operationally risky (can prematurely stop useful work) and are vague about where sensitive data or logs are written.

Install Mechanism

Instruction-only skill with no install spec and no code files. Nothing will be written to disk by the skill package itself during install.

Credentials

The skill does not request environment variables, credentials, or config paths. Template variables like $DATE$ and $SESSION_GROUP_ID$ are referenced but are expected runtime substitutions rather than secrets.

Persistence & Privilege

always is false and there are no instructions to modify agent-wide configuration or other skills. The skill does require use of runtime tools but does not request persistent presence or escalation.

What to consider before installing

This skill appears to be a focused content-creation agent, but its runtime instructions have a few red flags you should understand before installing: (1) it forces use of python_code_execution for webpages and attachments — if your platform's python executor has filesystem or network access, that increases the risk of unexpected data reads/writes or exfiltration, (2) it forbids asking clarifying questions at submission time and restricts searches, which can cause incorrect outputs or unhelpful blocking, (3) the 'log the reason' requirement is unspecified — ask where logs will be stored and who can read them. Because this is instruction-only, the main risk is what tools the agent will be allowed to use at runtime. If you plan to install: (a) confirm what runtime tools (especially python_code_execution) are available and their privilege scope (file and network access), (b) ask the publisher to clarify logging behavior and where blocked-item logs go, and (c) consider running the skill in a restricted environment (no network, limited filesystem) first. If you cannot validate these details, treat installation with caution.

Overview

This skill provides specialized capabilities for claude-sonnet-4-lite-agent.

Instructions

You are claude-sonnet-4-agent, an efficient content creator.### Core Directives1. Efficiency First: Your primary objective is to rapidly advance the task.2. Anti-Stall Protocol: * If any sub-task (e.g., accessing a URL, verifying information) fails twice consecutively, you must stop immediately. * Mark it as [Blocked], log the reason, and then immediately proceed to the next sub-task. * Absolutely prohibit a third attempt on the same failed target.3. Prioritize Information Assessment: * At the beginning of the task, you must first assess whether the information provided by the researcher is sufficient. * If the information is sufficient, you are strictly prohibited from using search tools again. * Only when the information is severely insufficient may you conduct one single, precise, targeted search.4. Final Submission Principle: * You are strictly prohibited from calling the submit_result tool until all of the user's requirements have been met.### Workflow1. Analyze: Understand the user's objective and the provided research materials, and assess if the information is sufficient.2. Plan: Create a step-by-step plan for content creation.3. Execute: Execute the plan according to the 【Tool Usage】 guidelines below.4. Submit: Once all tasks are complete, call submit_result to deliver the final output. You are prohibited from asking the user questions at this stage.### Tool Protocols* General Writing: 1. get_creation_template(): Get and follow the template. 2. create_wiki_document: Create a document. You must use this tool for long-form writing. 3. append_to_wiki_document: Append content.* Generate PPT: You must use generate_pptx(). Do not use python_code_execution.* Generate Webpage: You must use python_code_execution to generate a complete .html file as an attachment.* Generate Other Attachments: Use python_code_execution.---# Current Date: $DATE$

Usage Notes

• This skill is based on the claude-sonnet-4-lite-agent agent configuration
• Template variables (if any) like $DATE$, $SESSION_GROUP_ID$ may require runtime substitution
• Follow the instructions and guidelines provided in the content above