Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/benkapner/claude-code-basecamp/prompt-testgit clone --depth 1 https://github.com/Benkapner/claude-code-basecampWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/benkapner/claude-code-basecamp/prompt-test)<a href="https://agentmods.dev/commands/benkapner/claude-code-basecamp/prompt-test"><img src="https://agentmods.dev/badge/commands/benkapner/claude-code-basecamp/prompt-test.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00028 | $0.01156 |
| Opus 5 | $0.00014 | $0.00578 |
| Sonnet 5 | $0.00006 | $0.00231 |
| Haiku 4.5 | $0.00003 | $0.00116 |
Grade A, and why
prompt-test scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Prompt Test
Test your LLM prompts before deploying them. Run a prompt against sample inputs, inspect the outputs, and catch regressions when you modify prompts.
When to Use
- After editing a prompt template (e.g.,
prompts/person_prompts.py) - Before committing prompt changes to make sure outputs didn't degrade
- When comparing two prompt versions side-by-side
- When building a new prompt from scratch and iterating on quality
Instructions
1. Identify the Prompt
- If $ARGUMENTS specifies a file, read it and find the prompt function(s)
- If $ARGUMENTS is empty, search the current repo for prompt files (look in
prompts/, or files with "prompt" in the name) - List the available prompt functions and ask the user which to test
2. Prepare Test Inputs
- Look for existing test data in the repo (e.g., sample JSON files in
data/, test fixtures intests/) - If no test data exists, generate 2-3 realistic sample inputs based on the prompt's expected input format
- Show the user the test inputs and ask if they want to modify them
3. Build and Run the Prompt
- Call the prompt function with each test input to generate the actual prompt text
- Display the rendered prompt for each input so the user can review what's being sent to the LLM
- If the user wants to actually call the LLM (requires API key):
- Run the prompt against the configured LLM (check .env for GOOGLE_API_KEY, OPENAI_API_KEY, etc.)
- Show the raw LLM response for each input
- If the pipeline has a parsing function (e.g.,
parse_person_summary), run it on the response and show the parsed result
4. Evaluate Output Quality
For each test output, check:
- Completeness: Did the LLM fill all expected fields? Any empty or "Unknown" values that shouldn't be?
- Format compliance: Does the output match the expected format (e.g., SUMMARY/PRODUCTS/TECHNOLOGIES/WORK_FOCUS)?
- Grounding: Does the output reference specific details from the input, or is it generic/hallucinated?
- Length: Is the output within expected bounds (not too short, not too long)?
- Parseability: If there's a parsing function, does it successfully parse the output without falling back to defaults?
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 118 lines · 28 tokens per session scan A d7ff04a9c78d
prompt-test is a command published in the GitHub repository Benkapner/claude-code-basecamp (16 stars, last pushed 14d ago), licensed MIT. It adds 28 tokens to every session and 1,156 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
devkit.prompt-optimize
Provides expert prompt optimization using advanced techniques (CoT, few-shot, constitutional AI) for LLM performance enhancement. Use when you need to improve prompt quality or optimize LLM interactions.
prompt
System instructions for writing effective prompts. Apply when generating commands, skills, agents, or any LLM instructions.
create-meta-prompt
Create optimized prompts for Claude-to-Claude pipelines (research -> plan -> implement).
metaprompt
Generate a complete, ready-to-use prompt engineered for a specific model and harness.
prompt
Convert a rough ask into a structured XML-tagged prompt using the factory-prompting vocabulary.
guard
Analyze and prepare prompts with JSON for safe context submission.