메뉴
Multi-purpose utility for CRUX compression workflows. Provides token estimation and checksum calculation. Use when estimating tokens for compression, comparing file sizes, or getting checksums for sourceChecksum tracking.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Guided workflow for launching and interpreting the cursor-top TUI. Covers process discovery, hierarchy expansion, log tail interpretation, and recommended flags for different scenarios. Use when the user wants to monitor every Cursor agent on their machine in real time, or to capture a snapshot for a report.
Writes .zoto/eval-system/config.yml from field values pre-collected by /z-eval-configure (command-owned askQuestion). Refuses bundled update.preserveUserAuthoredCases:false or update.writeMetaMarker:false (including nested payload mirrors) before any manifest read or config write. Validates against templates/schema/config.schema.json. Diffs the chosen config against the manifest snapshot (discovery_config.static.framework) to produce a cleanup_plan validated against templates/schema/cleanup-plan.schema.json, and stamps _meta.primitive_analysis.invalidate=true on every cached primitive analysis when the static framework changes. The unified LLM eval harness ships one JSON-first co-located eval per target, with no per-repo selector axis — only the static framework matters here. Never calls askQuestion — returns needs_user_input if validation cannot proceed without user clarification.
Scaffolds static pytest and LLM (@cursor/sdk) eval backends into a host repository, generates eval cases with _meta.generated markers for approved skills/commands/agents/hooks, merges eval scripts into package.json, and writes the persistent manifest plus append-only history. Expects the invoking command to pre-collect approved targets — does not call askQuestion.
Invocation reference for all eval-system CLI commands. Use this skill when you need to run eval scripts — never reference script paths directly; always invoke via the package.json aliases documented here.
Executes eval runs via the host repo's package.json scripts (pnpm run eval, eval:full, eval:llm, eval:judge). Forwards --model to the LLM runner via both a CLI flag and the ZOTO_EVAL_MODEL environment variable. After each run, calls eval:update --check and appends the drift status to the run report as a warn-only line. Does not call askQuestion — credential gaps use needs_user_input for the command.
Adversarial judge for eval runs. Identifies weak graders, under-covered assertions, verbosity spikes, and accuracy/confidence regressions. Produces an enriched llm.yml with soft-metric annotations. Does not call askQuestion — handoff to /z-eval-update is expressed as needs_user_input for the command to confirm via askQuestion and resume. Uses the configured judgeModel (default claude-opus-4-8[]).
| name | crux-utils |
| description | Multi-purpose utility for CRUX compression workflows. Provides token estimation and checksum calculation. Use when estimating tokens for compression, comparing file sizes, or getting checksums for sourceChecksum tracking. |
Multi-purpose utility for CRUX compression workflows. Provides deterministic tools for token estimation and checksum calculation.
# Token count estimation
python3 .cursor/skills/crux-utils/scripts/crux-utils.py --token-count <file>
# Compare source vs CRUX (ratio mode, default target 25%)
python3 .cursor/skills/crux-utils/scripts/crux-utils.py --token-count --ratio <source> <crux>
# Compare with custom target (e.g. 40%)
python3 .cursor/skills/crux-utils/scripts/crux-utils.py --token-count --ratio <source> <crux> --target 40
# Get checksum for sourceChecksum frontmatter
python3 .cursor/skills/crux-utils/scripts/crux-utils.py --cksum <file>
--token-count <file>Estimate token count for a single file.
Output:
=== Token Estimate: example.md ===
Prose tokens: 397
Code tokens: 0
Special tokens: 0
---
TOTAL TOKENS: 397
--token-count --ratio <source> <crux> [--target <n>]Compare source file vs CRUX file and calculate compression ratio.
The --target option sets the compression target percentage (1-100). Defaults to 25 if omitted, matching the CRUX spec's recommended target for text/code content.
Output:
=== Compression Ratio Analysis ===
=== Token Estimate: source.md ===
...
=== Token Estimate: source.crux.md ===
...
=== Compression Summary ===
Source tokens: 397
CRUX tokens: 140
Ratio: 35.2% of original
Reduction: 64.8%
Target (≤25%): NO
--cksum <file>Get checksum of a file, formatted for CRUX frontmatter sourceChecksum field.
Output:
=== Checksum: example.md ===
Checksum: 1234567890
---
FRONTMATTER: "1234567890"
| Content Type | Chars/Token | Notes |
|---|---|---|
| Prose (markdown) | 4.0 | English text, headers, lists |
| Code blocks | 3.5 | More symbols, shorter identifiers |
| Special chars | 1.0 | CRUX Unicode symbols |
CRUX delimiters and symbols that count as 1 token each:
« » ⟨ ⟩→ ← ≻ ≺⊤ ⊥ ∀ ∃ ¬ ∋⊳ ⊲≥ ≤ ≠Δ Ρ Λ Π Κ Γ Φ Ω⊛ ◊ θBoth modes produce identical output for identical input:
For compression tasks:
--cksum <source> → use FRONTMATTER value for sourceChecksum--token-count <source> → use for beforeTokens--token-count <crux> → use for afterTokens--token-count --ratio <source> <crux> → check Target (≤25%)
--token-count --ratio <source> <crux> --target 40