Token Efficient Agent
v1.0.0Advanced techniques for minimizing token consumption in OpenClaw operations while maintaining or improving response quality. Includes memory optimization, do...
⭐ 0· 107·0 current·0 all-time
by@foinbo
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name, description, and the SKILL.md content align: guidance focuses on minimizing token usage via memory querying, document fetching, and tool-call strategies. There are no unrelated required env vars, binaries, or installs.
Instruction Scope
Instructions direct the agent to use memory_search/memory_get and document fetch APIs (e.g., feishu_fetch_doc) with precise offsets and summaries. This stays within the stated scope but explicitly instructs accessing user memory files and external documents, which is expected for the purpose but is privacy-sensitive. The guidance gives the agent discretion about when to escalate from summaries to full fetches—this is functional but means the agent may read user data during use.
Install Mechanism
Instruction-only skill with no install spec and no code files; lowest disk/write risk. Nothing is downloaded or installed.
Credentials
No environment variables, credentials, or config paths are requested by the skill. However, the techniques rely on the agent's access to memory and document tools (and thus to whatever credentials/permissions those tools use). That access is proportional to the skill's purpose but is sensitive because it reads user data.
Persistence & Privilege
always:false and normal invocation flags. The skill does not request persistent presence or modify other skills; it is user-invocable and can be called autonomously per platform defaults (not a unique privilege).
Assessment
This skill appears to be what it claims: a set of runtime instructions for reducing token usage. Before enabling or using it, consider: (1) Tool and permission scope — the skill explicitly tells the agent to read memories and fetch documents (feishu_fetch_doc/memory_get). Make sure the agent's tool permissions align with what you want it to access. (2) Privacy risk — techniques increase automated access to user data; review logs or run in a limited/test environment first. (3) Autonomy — the skill can be invoked by the agent as usual; if you want to avoid automated reads, restrict when or how the agent may call it. (4) Because it is instruction-only, no code is installed on disk, but the agent will still perform API/tool calls at runtime. If you don't want the agent to access memories or documents, do not enable this skill.Like a lobster shell, security has layers — review code before you run it.
latestvk97f4cb9433mzv495fqffb10jh8364hg
License
MIT-0
Free to use, modify, and redistribute. No attribution required.