determinism
Use when verifying outcomes with code instead of LLM judgment, versioning prompts with hashes, or ensuring reproducible agent behavior. Load for any critical verification. Scripts return boolean exit codes, not subjective assessments. Prompts use semantic versioning with SHA256 validation.
Packaged view
This page reorganizes the original catalog entry around fit, installability, and workflow context first. The original raw source lives below.
Install command
npx @skill-hub/cli install ingpoc-skills-determinism
Repository
Skill path: determinism
Use when verifying outcomes with code instead of LLM judgment, versioning prompts with hashes, or ensuring reproducible agent behavior. Load for any critical verification. Scripts return boolean exit codes, not subjective assessments. Prompts use semantic versioning with SHA256 validation.
Open repositoryBest for
Primary workflow: Ship Full Stack.
Technical facets: Full Stack, Testing.
Target audience: everyone.
License: Unknown.
Original source
Catalog source: SkillHub Club.
Repository owner: ingpoc.
This is still a mirrored public skill entry. Review the repository before installing into production workflows.
What it helps with
- Install determinism into Claude Code, Codex CLI, Gemini CLI, or OpenCode workflows
- Review https://github.com/ingpoc/SKILLS before adding determinism to shared team environments
- Use determinism for development workflows
Works across
Favorites: 0.
Sub-skills: 0.
Aggregator: No.
Original source / Raw SKILL.md
---
name: determinism
description: "Use when verifying outcomes with code instead of LLM judgment, versioning prompts with hashes, or ensuring reproducible agent behavior. Load for any critical verification. Scripts return boolean exit codes, not subjective assessments. Prompts use semantic versioning with SHA256 validation."
keywords: verification, deterministic, hash, version, reproducible, code-not-judgment
---
# Determinism
Reproducible outcomes through code verification and prompt versioning.
## Core Principle
> "Claude can run scripts without loading either the script or the PDF into context. And because code is deterministic, this workflow is consistent and repeatable." - Anthropic Engineering
## Instructions
1. Replace LLM judgment with script verification
2. Version prompts with semantic versioning
3. Hash-validate critical prompts: `scripts/validate-prompt.sh`
4. Use exit codes (0 = pass, 1 = fail), not text
## LLM Judgment vs Code Verification
| Task | LLM (Bad) | Code (Good) |
|------|-----------|-------------|
| Tests passed? | "The tests appear to pass" | `pytest; echo $?` → 0 or 1 |
| Valid JSON? | "This looks like valid JSON" | `python -c "json.load(f)"` |
| Server running? | "The server should be up" | `curl -s localhost/health` |
## References
| File | Load When |
|------|-----------|
| references/code-verification.md | Writing verification scripts |
| references/prompt-versioning.md | Versioning/hashing prompts |
---
## Referenced Files
> The following files are referenced in this skill and included for context.
### scripts/validate-prompt.sh
```bash
#!/bin/bash
# Validate prompt hash matches declared hash
# Usage: validate-prompt.sh PROMPT_FILE
# Exit: 0 = valid, 1 = invalid/no hash
PROMPT_FILE=${1:-"PROMPT.md"}
if [ ! -f "$PROMPT_FILE" ]; then
echo "ERROR: File not found: $PROMPT_FILE"
exit 1
fi
# Extract declared hash from header
DECLARED=$(grep -o 'SHA256: [a-f0-9]\+' "$PROMPT_FILE" | head -1 | cut -d' ' -f2)
if [ -z "$DECLARED" ]; then
echo "WARN: No hash declared in $PROMPT_FILE"
exit 0 # No hash = no validation (allow)
fi
# Calculate actual hash (strip comment lines)
CONTENT=$(grep -v '^<!--' "$PROMPT_FILE" | grep -v '^$')
ACTUAL=$(echo "$CONTENT" | shasum -a 256 | cut -d' ' -f1)
# Compare (declared may be truncated)
if [[ "$ACTUAL" == "$DECLARED"* ]]; then
echo "PASS: Hash valid for $PROMPT_FILE"
exit 0
else
echo "FAIL: Hash mismatch in $PROMPT_FILE"
echo " Declared: $DECLARED"
echo " Actual: ${ACTUAL:0:32}..."
exit 1
fi
```