saross
10 verified skills
improve-prompt
Systematically improve a seed prompt using anti-satisficing techniques and structured prompt hardening. This skill should be used when the user wants to strengthen, harden, or optimise a prompt before deploying it — particularly for review, comparison, or analytical tasks where LLMs tend to satisfice. Trigger phrases include "improve this prompt," "harden this prompt," "make this anti-satisficing," or `/improve-prompt [seed text]`.
testing
review-implementation
Structured review protocol for catching suboptimal implementations and methodology choices. Use when completing a significant new integration, adopting a new tool or method, or at phase boundaries in a project. Surfaces both 'discovery failures' (capabilities not yet considered) and 'exploitation failures' (capabilities known but underutilised). Invoke with /review-implementation.
tools
skills/entity-classifier
Classify dual-nature entities (hotels, churches, schools, halls) as building-only, business/organisation-only, or both based on contextual linguistic analysis.
development
field-type-docs
Generate modular "lego brick" documentation for Fieldmark field types. Produces design docs (Notebook Editor configuration), collect docs (data collection usage), shared docs, Playwright screenshot specs, and practical fieldwork tips. This skill should be used when creating, updating, or reviewing field type documentation for the fieldmark-docs-staging repository.
development
phase-gate
Experimental phase boundary checkpoint. Use before committing API spend, compute, or time to a new experimental phase — especially when scaling from pilot to production, applying a configuration selected on a subset to a larger dataset, or transitioning between optimisation and evaluation phases. Surfaces under-powered assumptions from prior phases and determines whether cheap validation is needed before proceeding. Trigger phrases: 'ready to run', 'start Phase X', 'scale up', 'production run', 'let's proceed with'. Also trigger PROACTIVELY when recognising a phase boundary, even if the user does not invoke it explicitly.
development
reflect
End-of-session reflection protocol. Use when the user invokes /reflect or asks to reflect on the current session, do end-of-session reflections, or similar. Guides Claude through updating reflection documents in priority order, answering structured prompts, and maintaining the reflection and investigation logs.
documentation
notebook-creator
Generate valid Fieldmark notebook JSON files from natural language descriptions, field manuals, or specifications. Supports validation rules, conditional logic, and parent-child relationships.
testing
build-rubric
This skill should be used when the user asks to "build a rubric", "create a rubric", "design a rubric", "make a marking rubric", "assessment rubric", or "rubric for [assessment name]". Also trigger when the user mentions needing to mark or assess student work and a rubric does not yet exist. Do not trigger for rubric review or revision of an existing rubric — only for building a new one.
development
moderate-mark
This skill should be used when the user asks to "moderate marks", "produce marking dossiers", "double-mark" an assessment, run a "second-reader pass", or "build a moderation pack". Also trigger when the user has just entered rubric marks for a HUMN8031 Assessment 2 paper and wants a moderation dossier produced. Do not trigger for rubric design or rubric review — only for dossier production on a marked assessment.
development
audit-config
Pre-launch experimental configuration audit. Use before committing API spend to a detection or verification run. Systematically checks every config parameter against the preregistered protocol, known failure modes, and the filesystem. Produces a READY TO LAUNCH / BLOCKED verdict. Trigger phrases: 'audit configs', 'check configs', 'ready to run', 'launch the evaluation'. Also trigger PROACTIVELY when the user is about to launch an experimental API run — especially after generating new configs.
development