optimize-search-runtime
/optimize-search-runtime
Select and execute the smallest sufficient OreoLook search plan. Use whenever a request may need current web evidence, source fetching, or deep research and runtime cost, latency, context size, tool count, and output depth must remain bounded.
- Author
- elixpo ↗
- Repository
- elixpo_chapter ↗
- Licence
- MIT
- Last repository change
- 27.10.2026
Assessment and security
- Files in this package
- 2
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 27.10.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Select and execute the smallest sufficient OreoLook search plan. Use whenever a request may need current web evidence, source fetching, or deep research and runtime cost, latency, context size, tool count, and output depth must remain bounded.
Original author description
Select and execute the smallest sufficient OreoLook search plan. Use whenever a request may need current web evidence, source fetching, or deep research and runtime cost, latency, context size, tool count, and output depth must remain bounded.
How it works
- Plan
- Context
- Output
- Runtime contract
SKILL.md preview
---
name: optimize-search-runtime
description: Select and execute the smallest sufficient OreoLook search plan. Use whenever a request may need current web evidence, source fetching, or deep research and runtime cost, latency, context size, tool count, and output depth must remain bounded.
---
# Optimize Search Runtime
Choose depth from meaning and required evidence, never keyword matching.
## Plan
1. Use `quick` for one live fact or a narrow lookup: at most 1 search, 2 fetched sources, and 350 answer tokens.
2. Use `standard` for one subject requiring verification or comparison: at most 2 searches, 4 fetched sources, and 700 answer tokens.
3. Use `deep` only for a genuinely multi-part investigation: at most 4 subqueries, 2 iterations per subquery, 4 fetched sources per subquery, and 1,800 final tokens.
4. Reuse cached evidence and conversation summaries before requesting new work.
5. Stop as soon as the evidence supports the answer. Never spend remaining budget merely because it exists.
## Context
- Include only history needed to resolve references or maintain continuity.
- Prefer the durable summary over raw older turns.
- Keep ordinary injected history within 3,000 tokens and deep history within 6,000 tokens.
- Keep retrieved semantic context within 4,000 characters and the five most relevant memories.
## Output
- Answer first and cite only sources actually used.
- Avoid repeating tool evidence or narrating internal work.
- Match response length to the request; a quick lookup should remain quick.
## Runtime contract
agent: policy
tools: []
timeout_seconds: 1
max_concurrency: 1
output: bounded_search_policy
Files in this package (2)
- SKILL.md 2 KB
- agents/openai.yaml 1 KB
Versions
The database has no other skills named optimize-search-runtime.
Database snapshot: 13.11.2026 · package version 4fec0dfafa8d (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 100web-research
Expert web research router. Automatically selects the best search and page-fetch tool for any research task. Invoke whenever performing web search, fetching documentation, looking up library APIs, or retrieving any external web content. The user never needs to specify which tool… - 100model-router
Recommend which model or provider preset to use based on cost, latency, context size, and task difficulty. Use when the user asks which model to pick or when work is mismatched to the current model. - 91web-search-backend-configuration
Use for configuring web search backend for one-off user queries. - 90context-engineering
Engineer what enters the model's context window for reliability and cost using Write, Select, Compress, and Isolate. Use when: context engineering, context window budget, context rot, prompt caching, just-in-time retrieval, context compaction, sub-agent context isolation, tool output distillation. - 91context-architect
Context engineering tool for building AI agent systems. Use when developers need to design agent context strategies — system prompts, tool sets, examples, and runtime context management. Guides through modular design with built-in patterns and dynamic research. - 100router-tool-selector
Use to select the minimal toolset required to fulfill a request. Defaults to no tools and adds tools only when there is a clear, specific reason — preventing unnecessary RAG retrieval, web search, or paid API calls that add latency and cost without improving response quality.…