guard-the-context-window

Files (1)

• SKILL.md

SKILL.md (first lines)

# Guard the Context Window

The context window is finite and non-renewable within a session. Every token that enters should earn its place. Models reason measurably worse as the window fills; a focused context is a capability, not a nicety.

**Scope:** this applies to long-running work — multi-file implementation, orchestration, wiki generation, verification sweeps. A quick question or a back-and-forth conversation answers directly; routing a simple lookup through a subagent wastes time and adds nothing.

**Why:** Context overflow degrades reasoning quality, creates compression artifacts, and halts progress. Unlike compute or time, context spent inside a session cannot be reclaimed.

**Pattern:**
- **Isolate large payloads.** Route verbose outputs, screenshots, and large documents to subagents. The main context gets summaries, not raw data.
- **Don't read what you won't use.** Read selectively based on relevance. If a file isn't needed for the current task, skip it.
- **Keep frequently used content inline.** Templates and references used on every invocation belong in the skill file, not in separate files that cost a read each time.
- **Size phases and cap scope.** Limit files per phase, set turn budgets, account for mechanism costs.

From baranbingol1/bbstack, MIT.

More from baranbingol1/bbstack