AB predictalot
Self-hosted forecasting/prediction service. Foundation time-series endpoints under /v1/timeseries/<type>/{forecast,forecast/ensemble} + GET .../models — univariate, past/future/both covariates, multivariate, samples — over 5 zero-shot models (chronos-2, timesfm-2.5, moirai-2, toto-1, sundial-base-128m). Plus supervised tabular ML under /v1/tabular/ (9 backends — lightgbm, xgboost, hist-gbt, random-forest, logistic, mlp, svm-rbf, knn, naive-bayes — over direction/value/quantile modes) with train+persist, weighted ensembles, and calibrated/stacking/diversified meta-learners. Unified REST + MCP (streamable-HTTP at /mcp, one tool per (type, model) cell + per-type ensemble + listing) + optional bearer auth. Use when the user wants to forecast a numeric time series (quantile bands or raw sample paths), condition a forecast on known/future covariates, ensemble several forecasters, or train a tabular model on engineered features and predict direction/value/quantiles on the latest snapshot.
Self-hosted forecasting/prediction service.
As a process B 66/100 · Nearly there — weak spots: inputs and preconditions, execution cost, running it twice
How to improve
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 4. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 8046 tokens (recommended < 5000); move details to references/ - note
frontmatter-keyunknown frontmatter key "homepage" - note
frontmatter-keyunknown frontmatter key "permissions"
Process rating: all ten parameters 66/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 22 mutating operations with no state check
- 40Execution cost. Instruction body is 8046 tokens: crowds the task out of the window
- 60Tools and files. Uses tools (web) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 85Steps. 40 steps, 1 vague phrases
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 1 branches, has a failure section
- 100Consistency. Name and required fields are in place
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 21 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (15 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 996: 120–800 characters recommended
- -2localhost URLs: will not work for another user
- +1No license
- +2Single-language instructions
- +4Structure: 53 headings
- +3Step-by-step instructions: 40 items
- +3Output format is stated explicitly
- +4Has examples (16 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
- +3All 1 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 77.