dhpk-harness-budgetlisted

# Harness Budget (context-window accounting) Analyze token overhead across every loaded component in a Claude Code session and surface actionable optimizations to reclaim context space. Resolve `HARNESS_DIR` using `@skills/dhpk-harness-revise/references/harness-directory-contract.md` before counting harness components. > See also: `/harness-govern` chains this token audit into an end-to-end harness governance loop. ## When to Use - Session performance feels sluggish or output quality is degrading - You've recently added many skills, agents, or MCP servers - You want to know how much context headroom you actually have - Planning to add more components and need to know if there's room - Running `/dhpk:dhpk-harness-budget` command (this skill backs it) ## When NOT to Use - Trimming, deduplicating, or validating harness structure — use `dhpk-harness-revise`. - Backfilling missing `.claude/` skills / agents / rules — use `dhpk-harness-fill`. - Auditing `.claude/` config health, naming, or hook wiring — use `/dhpk:dhpk-claude-health`. - Reviewing business code — route through the reviewer agents. ## How It Works ### Phase 1: Inventory Scan all component directories and estimate token consumption: **Agents** (`agents/*.md`) - Count lines and tokens per file (words × 1.3) - Extract `description` frontmatter length - Extract `model:` + `effort:` frontmatter and the tool list (feeds the Phase 3b tier-economics audit) - Flag: files >200 lines (heavy), description >30 words (blo