Typesafe Calibrator — Development agent for Claude Code
Turns invented thresholds into measured ones.
How to install Typesafe Calibrator
Installs to ~/.claude/agents/herakles-dev-typesafe-claude-kit-typesafe-calibrator.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/herakles-dev/typesafe-claude-kit/HEAD/agents/typesafe-calibrator.md -o ~/.claude/agents/herakles-dev-typesafe-claude-kit-typesafe-calibrator.md Restart Claude Code, or start a new session, for it to be picked up.
What Typesafe Calibrator does
name: typesafe-calibrator description: "Turns invented thresholds into measured ones. Builds labelled sets, plots confidence against accuracy, measures self-consistency across repeated runs, and sets the confidence gates that let a TypeSafe decision act automatically. Required before any judgment gates an irreversible action." model: sonnet color: yellow category: typesafe default_mode: subagent effort: high triggers:
- "validate the thresholds"
- "calibrate"
- "what confidence cutoff"
Alternatives in Development
- Compass Planner — Use this agent when the user needs a structured work plan from a vague idea, when they say 'plan this' or 'let 523 ★
- Sub Pricing — Internal subagent 132 ★
- Po Leverage — Leverage seat on the Atlas PO Council 79 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Dr Efficiency
Deep-review finder angle. Finds wasted work the diff introduces - repeated expensive construction, re-parsing,
Gad Signal Lead
Develops event selection criteria, optimizes signal region definition, performs categorization studies, and pr
Scientific Figures
Generate publication-quality scientific figures with consistent styling. Use when experiments need visualizati
Extractor Plots
Extract story elements from document chunks (quests, scenes, themes, plot points)
Transcript Proofer
Proof-read an ASR words.json against the media before captions are written. Re-runs ASR on suspect spans and r
Perf Engineer
Performance diagnosis specialist for the Hydraia pipeline. Measurement-first — establishes a baseline, profile
Related Skills
Jev Search Rerank Eval
Does a TypeSafe Jev rerank beat embedding search? Graded relevance eval (9,831 pairs, 164 zh/en queries) over
Typesafe Jev
Field notes, runnable scripts and an agent skill for TypeSafe Jev, the System One decision model. Measured eva
Videoclip
Run the full music-video pipeline from one clip of the artist singing and one track. Locks the artist, measure