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.
git clone --depth 1 https://github.com/katalon-labs/true-skillsWrote 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/katalon-labs/true-skills/test-estimation)<a href="https://agentmods.dev/commands/katalon-labs/true-skills/test-estimation"><img src="https://agentmods.dev/badge/commands/katalon-labs/true-skills/test-estimation/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/katalon-labs/true-skills/test-estimation"><img src="https://agentmods.dev/badge/commands/katalon-labs/true-skills/test-estimation.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00194 | $0.06986 |
| Opus 5 | $0.00097 | $0.03493 |
| Sonnet 5 | $0.00039 | $0.01397 |
| Haiku 4.5 | $0.00019 | $0.00699 |
Grade A, and why
test-estimation 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 yesterday.
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 — 405 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Katalon Test Estimation
Use this skill for the estimation half of the planning stage: turn a scoped cycle into person-hours, a date, and a headcount. The output is a range with a confidence label, never a single number. test-plan decides what gets tested; this skill decides what that costs.
Availability Boundary
State this before producing any number.
The Katalon MCP can count the work. It cannot measure the effort.
- Available via MCP: case and suite counts (
find_test_cases,read_test_suite,find_test_cases_by_requirement), uncovered requirements (fetch_requirement_data), automated/manual split when exposed (fetch_test_case_data), failure and flake rates (fetch_test_stability_data,find_test_results), the configuration matrix (fetch_test_configuration_data), the cycle window (find_iterations), and a past execution's machine-recorded elapsed time (read_execution,read_execution_test_results). - Not available at all: how long a person takes. No tool books effort against a case, a run, or a person. A manual run's open and close timestamps are calendar elapsed, not work.
- Also not available: team roster, capacity, PTO, or working calendar; any cost or rate model; a field to write the estimate back into; any cross-cycle aggregate such as average cycle duration.
- Looks like MCP data but is not: session concurrency.
list_test_cloud_environmentsandfind_execution_profileslist environments and profiles, not the number of parallel sessions the plan allows. Tag it team-supplied. - Field-level rule: read one record with
read_executionand confirm the elapsed-time field is present before modelling on it. If it is absent, ask for the figure and label it team-supplied. Never let a missing field become a zero.
The consequence: every per-unit rate is team-supplied and must be tagged as such in the output. Read references/mcp-evidence-map.md for the full signal-to-tool map and the human-input checklist.
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.
- yesterday First seen · 405 lines · 194 tokens per session scan A 5d89e5e72d06
test-estimation is a command published in the GitHub repository katalon-labs/true-skills (8 stars, last pushed 2d ago), licensed MIT. It adds 194 tokens to every session and 6,986 once invoked, about $0.0010 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 2026-09-08.
Other commands, from other repositories
implement
Execute tasks from a track's implementation plan following TDD workflow.
test-review
Act as a software quality engineer and python expert.
run_tests
Use the text the user typed after the command name as the request for this command.
write-unit-tests
Create comprehensive unit tests for the current code and generate the test file with proper imports and setup according to the project's testing conventions.
test
Command "test" from duongductrong/cursor-kit, covering purpose, analyze, test categories, output and rule.
implement_tests
Use the text the user typed after the command name as the request for this command.