← Back to Skills Marketplace
132
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install skill-retrieval-gate
Description
Decide whether to run `memory_search` before following another skill or workflow, so the agent can reduce token usage without forcing retrieval on every task...
Usage Guidance
This skill is a lightweight decision-layer that tells the agent whether to call your memory_search. It does not itself access credentials or external endpoints. Before installing, confirm your agent actually exposes a safe memory_search implementation (so queries don’t leak sensitive data to unintended sinks), and decide whether you want the agent to invoke retrieval autonomously. Also periodically review the decision rules to ensure they fit your workflows (to avoid unnecessary retrievals or skipped context).
Capability Analysis
Type: OpenClaw Skill
Name: skill-retrieval-gate
Version: 1.0.0
The skill bundle consists entirely of Markdown-based instructions designed to help an AI agent optimize its use of a `memory_search` tool. It provides logical frameworks for decision-making, query construction, and result trimming to reduce token usage and improve context relevance. There is no executable code, no evidence of data exfiltration, and no malicious prompt injection; the instructions in SKILL.md and the references directory are strictly aligned with the stated purpose of retrieval management.
Capability Assessment
Purpose & Capability
Name/description promise (decide whether to run memory_search) matches the included instructions and reference files. No unrelated env vars, binaries, or capabilities are requested.
Instruction Scope
SKILL.md and reference docs only provide rules for deciding when/how to call memory_search, how to build queries, trim results, and when to fall back. They do not instruct reading system files, contacting external endpoints, or exfiltrating data.
Install Mechanism
Instruction-only skill with no install spec and no code files; nothing is written to disk or downloaded.
Credentials
No environment variables, credentials, or config paths are required. The guidance assumes an existing memory_search capability but does not demand any secrets.
Persistence & Privilege
always is false and there are no indications the skill will modify other skills or agent-wide settings. Autonomous invocation is allowed (platform default) but the skill itself is low-privilege.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install skill-retrieval-gate - After installation, invoke the skill by name or use
/skill-retrieval-gate - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of skill-retrieval-gate.
- Enables intelligent decision-making on when to run `memory_search` before other skills or workflows.
- Aims to reduce unnecessary retrievals, minimizing token usage by judging real dependency on local knowledge, history, or prior decisions.
- Provides practical rules for when and how to retrieve, limiting context to top results, and emphasizing fast fallback if retrieval is weak or not needed.
- Clarifies common triggers, decision workflow, anti-patterns to avoid, and output expectations for agents.
Metadata
Frequently Asked Questions
What is skill-retrieval-gate?
Decide whether to run `memory_search` before following another skill or workflow, so the agent can reduce token usage without forcing retrieval on every task... It is an AI Agent Skill for Claude Code / OpenClaw, with 132 downloads so far.
How do I install skill-retrieval-gate?
Run "/install skill-retrieval-gate" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is skill-retrieval-gate free?
Yes, skill-retrieval-gate is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does skill-retrieval-gate support?
skill-retrieval-gate is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created skill-retrieval-gate?
It is built and maintained by WeiHan (@otweihan); the current version is v1.0.0.
More Skills