Token Saver
v0.1.0Reduce Claude API token consumption and track spending. Diagnoses waste patterns, recommends optimizations, and generates cost reports. Use when: (1) User as...
⭐ 0· 38·0 current·0 all-time
bynicobot@nicope
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The skill advertises cost analysis and optimizations for Claude/Anthropic usage and only requires ANTHROPIC_API_KEY and local OpenClaw config access, which is appropriate for fetching billing/usage and inspecting local agent configuration to recommend fixes.
Instruction Scope
SKILL.md explicitly instructs the agent to call the Anthropic usage endpoint, inspect local files (openclaw.json, SOUL.md, AGENTS.md, MEMORY.md), count session turns, and list installed skills. These actions are within scope for diagnosing token waste. The documentation also states it will show diffs and ask for confirmation before applying edits.
Install Mechanism
No install spec; skill is instruction-only with a small included Python script. No downloads or archive extraction are performed. The included script is readable and uses only the standard library.
Credentials
Only ANTHROPIC_API_KEY is required. That key is necessary to query the Anthropic usage API and is proportional to the skill's purpose. The skill does not request other unrelated credentials or sensitive environment variables.
Persistence & Privilege
always:false (default). The skill does not request permanent elevated presence and its docs explicitly require confirmation before editing local config files. No evidence it attempts to modify other skills or system-wide settings without user approval.
Assessment
This skill appears coherent with its stated goal. Before installing or running it: (1) inspect openclaw.json, MEMORY.md and other local files yourself and make backups—the skill will propose edits but you should verify them; (2) be aware that providing ANTHROPIC_API_KEY allows the script to query your usage and (depending on the key's privileges) possibly interact with Anthropic APIs—use a key with minimal necessary scope or rotate it after use if you have concerns; (3) run the included Python script locally (it uses only stdlib urllib) to see outputs and confirm it behaves as expected; (4) if you allow the skill to modify files, require explicit review of diffs before applying. If you want higher assurance, review the token_report.py source and the SKILL.md recommendations line-by-line (both are included) before use.Like a lobster shell, security has layers — review code before you run it.
latestvk97a3z82q0j384qh2jgw9jr5d983zq9s
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
💰 Clawdis
EnvANTHROPIC_API_KEY