Practitioner briefing
Reduce token spend, avoid rework, and verify what ships.
Need the full training? Open the 8-hour workshop.
Reduce unnecessary input, output, and tool overhead.
Keep context focused. Avoid expensive rework loops.
Plan, test, and use the right agent mode for the task.
Community guidance, not official GitHub or Microsoft policy.
System prompt, instructions, files, history, tool schemas, and your prompt.
Responses, reasoning, and tool calls. Usually the highest-cost token type.
Stable request prefixes can be reused at a lower effective cost.
Every read, command, retry, and rework step compounds cost.
Choose formats per task. Caveman is optional, not a guaranteed cost reduction.
Compare default, concise, and Caveman on total cost and correctness. Evidence and protocol.
Project pitfalls, required commands, naming, architecture rules, output constraints.
Boilerplate, facts visible in code, duplicate rules, occasional checklists.
Loaded instructions add input on later turns. Put path-specific rules in scoped files and occasional guidance on demand.
applyTo instruction files for path-specific rules.For one long task, keep this stable:
Changing these settings can lose cache savings. Start a fresh chat with a short handoff when a change is needed.
Questions, explanations, decisions. No tools needed.
Bounded changes with known files and intent.
Multi-file work that needs exploration, commands, and verification.
Do not use Agent mode for a question that Ask mode can answer.
Choose effort before long work. If the task changes, reassess with a concise handoff.
Capture acceptance criteria, likely files, risks, and verification. Start execution with only that handoff.
/context in Copilot CLI to see what is loaded.Compiled defaults for common developer command output.
YAML pipelines and local savings tracking.
Choose one filter layer per command path. Do not stack both by default.
Graphify builds a reusable repo map for structural queries. Update it after refactors.
graphify query "where is auth middleware?"
graphify path "Router" "Database"
Best for large repos and repeated agent sessions. Skip it for small, one-off changes.
Use /context to inspect Copilot CLI context pressure and loaded tools.
Tokentop shows local live tokens, cost, burn rate, and budget alerts.
Use these to find waste; neither tool compresses context.
Choose these settings before long work. If they need to change, start a fresh session with the relevant context.
Check the result before review exposes missing work.
Set concise output preferences.
Trim loaded instructions, files, and tools.
Try one output filter; preserve diagnostics.
Clarify the task, model choice, and verification.
Choose a change for observed waste. Compare accepted work and total cost.