Token Saver For Programmer

PassAudited by ClawScan on May 13, 2026.

Overview

This instruction-only skill is coherent and has no code, install, network, credential, or persistence behavior, but users should review its rewritten code or prompts because its “lossless” claims are qualified by its own warnings.

You can install this as a text-rewriting helper, but do not rely on it as truly lossless. Review and test any optimized code, configuration, logs, or prompts before applying them, especially in production or debugging workflows.

Publisher note

This version adds explicit safety notes in the README/SKILL.md to remind users to review outputs before use. The skill auto-activates on code/log/config/prompt inputs and performs token-saving minification. While core logic is preserved, edge cases may have subtle differences; users are advised to compare and test before production use. No network access is required.

Findings (2)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

NoteHigh Confidence
ASI01: Agent Goal Hijack
What this means

Condensed code, configs, logs, or prompts may lose nuance or change edge-case behavior.

Why it was flagged

The skill may automatically transform user-provided prompts or code. This is its stated purpose, but automatic rewriting can change meaning if the user applies the result without review.

Skill content
auto_activates: [code,log,config,prompt] ... This skill automatically compresses and rewrites code, logs, configs and prompts. It may omit subtle details or change conditional expressions in edge cases.
Recommendation

Compare the optimized output with the original and test code/config changes before using them in real projects.

What this means

A user might assume rewritten code or prompts are fully equivalent when they may not be.

Why it was flagged

The skill advertises lossless/no-logic-loss optimization while also warning that the output may not be strictly lossless. This internal qualification reduces concern, but users should not over-trust the marketing language.

Skill content
description: ... save 40%~60% tokens without losing logic ... Safe & lossless optimization ... Do not assume the transformation is strictly lossless.
Recommendation

Treat the token-saved result as a suggested rewrite, not a guaranteed equivalent transformation.