juxt

Track session intent and progress against GitHub Issues, and write every GitHub-bound prose body (issue, comment, description, progress update) in the chalk voice. Use when the user says "chalk

Write or update a technical documentation page in the chalk voice, structured around Diataxis. Use when the user says "tend the docs", "update the docs", "write a docs page", "document this feature", "add a how-to for X", "/chalk:tend-docs", or references adding/editing an end-user-facing docs page in a technical project.

Create a commit with a contextual body explaining the why, in the chalk voice. Use when the user says "commit this", "commit", "make a commit", "/chalk:commit"; OR is about to compose, draft, write or amend any git commit message body (e.g. "write a commit message", "draft the commit body", "amend the commit message"). Load this skill BEFORE drafting any such prose — it carries the voice guidance the commit body needs.

Create a pull request with a description that captures the intent and reasoning behind the change, in the chalk voice. Use when the user says "create a PR", "open a PR", "submit a PR", "raise a PR", "make a PR", "PR this", "/chalk:pr"; OR is about to compose, draft, write or update any pull request title or body (e.g. "write a PR description", "draft the PR body", "update the PR description", "let's put that in the PR"). Load this skill BEFORE drafting any such prose — it carries the voice guidance the PR body needs.

Build or query a code-only symbol and call-graph map of the current project, via LSP. Use when exploring an unfamiliar codebase, locating a symbol's definition, tracing what calls or is called by a function, assessing the blast radius of a proposed change, or navigating code beyond what grep alone can reveal. Produces a JSON map under .codemap/ at the project root. Skip for trivial lookups and single-file edits.

Audit technical documentation for drift against a code change. Produce a punch list of user-facing docs pages that probably need updating. Use when the user says "weed the docs", "check docs for drift", "audit docs against this diff", "which docs does this change affect", "/chalk:weed-docs", or is about to open a PR and wants a docs-impact check.

Interrupt forward momentum and step back to reassess

Give your AI agents something more useful than a prompt. Velocity through clarity.

Extract an Allium specification from an existing codebase. Use when the user has existing code and wants to distil behaviour into a spec, reverse engineer a specification from implementation, generate a spec from code, turn implementation into a behavioural specification, or document what a codebase does in Allium terms.

Run a structured discovery session to build an Allium specification through conversation. Use when the user wants to create a new spec from scratch, elicit or gather requirements, capture domain behaviour, specify a feature or system, define what a system should do, or is describing functionality and needs help shaping it into a specification.

Generate tests from Allium specifications. Use when the user wants to propagate tests, generate test files from a spec, write tests for a specification, create property-based tests, produce state machine tests, check test coverage against spec obligations, or understand what tests a specification requires.

Tend the Allium garden. Use when the user wants to write, edit, update, add to, improve, clarify, refine, restructure, fix or migrate Allium specs. Covers adding entities, rules, triggers, surfaces and contracts, fixing syntax or validation errors, renaming or refactoring within specs, migrating specs to a new language version, and translating requirements into well-formed specifications. Pushes back on vague requirements.

Weed the Allium garden. Find where Allium specifications and implementation code have diverged, and help resolve the divergences. Use when the user wants to check spec-code alignment, compare specs against implementation, audit for spec drift or violations, sync specs with code or code with specs, or verify whether the implementation matches what the spec says.

Use when about to run commands with potentially long output (>40 lines) — builds, tests, logs, kubectl, docker, grep, find. Pipes output through glance for token-efficient summaries instead of reading all tokens.