Save Usage ~4x (Codex oAuth users)
v1.0.0Classifies tasks by safety and complexity to route safe, negligible queries to gpt-5.1-codex-mini and escalates critical or executable tasks to gpt-5.3-codex.
⭐ 0· 567·3 current·3 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (route safe queries to a cheaper model, escalate risky/executable ones) matches the SKILL.md instructions. It requires no env vars, binaries, or installs — appropriate for a purely behavioral routing skill. The only implicit dependency is the platform-provided sessions_spawn() API, which is a reasonable requirement for a routing skill.
Instruction Scope
Instructions stay within routing scope: classify the task, forward the full task to sessions_spawn() when triggers match, and restrict what the cheaper model may produce. The skill will forward the user's full task to gpt-5.3-codex when escalating, which is expected but is a privacy/telemetry consideration (user content will be sent to the higher-capability model). The instructions do not ask the agent to read files, env vars, or system state outside of the task text.
Install Mechanism
No install spec or code files are present (instruction-only), so nothing is downloaded or written to disk — lowest-risk install footprint.
Credentials
The skill requests no environment variables, credentials, or config paths. That is proportional to its described behavior as a pure routing policy.
Persistence & Privilege
The skill is not always:true and does not request system-wide privileges. It does instruct the agent to escalate automatically on many common triggers; that autonomous invocation behavior is expected for a routing skill but has operational implications (frequent escalations can increase cost and send more user content to gpt-5.3-codex).
Assessment
This skill is coherent and does what it says: it classifies inputs and forwards risky/executable/high-impact tasks to a stronger model. Before installing, confirm: (1) your platform actually provides the sessions_spawn() API and you understand what it will send (it forwards the full task prompt to gpt-5.3-codex); (2) you're comfortable with potential privacy and billing implications of more frequent escalations; (3) you want the agent to make automatic escalation decisions rather than prompting you first. Test with non-sensitive prompts to see how often it escalates and monitor costs and logs. If you need stricter privacy, require explicit user confirmation before escalation or redact sensitive material before forwarding.Like a lobster shell, security has layers — review code before you run it.
latestvk97dvp0fkzysmfwtqw48yqgnxn815mz6
License
MIT-0
Free to use, modify, and redistribute. No attribution required.