atelier

Use when the user runs "/insights-audit", asks to "validate the insights report", "audit the report against my projects", "check what the report got wrong", or wants to cross-check an /insights-generated HTML report against actual GitHub repos and local project directories.

This skill should be used when the user asks to "end session", "capture state", "session summary", "what changed this session", "summarize repos", "write session notes", or wants to capture multi-repo state at the end of a work session.

Fill in the Patterns & Surprises section of the latest session reflect file and write the completed review to .ctx/logs/self-review/. Use when the user runs "/self-review", asks to "fill in the reflect", "write the session review", or "what patterns did we hit".

Use when working on the rslm project — Recursive Language Model inference engine in Rust. Triggers on "rslm", "RLM", "recursive language model", "rhai built-ins", "ctx_grep", "ctx_slice", "rlm_call", "rslm-core", "rslm-providers", "rslm-store", or any question about running queries, ingesting documents, or writing Rhai scripts for the RLM loop.

This skill should be used when the user asks to "review this diff", "ai review", "security review my changes", "review before push", "check my code for issues", or wants an LLM-based security and correctness review of uncommitted or unpushed changes.

This skill should be used when the user asks to "generate commit message", "write a commit msg", "ai commit", "generate a conventional commit", "write my commit", or wants an LLM to draft a commit message from the staged diff.

Use this skill when the user says "build a harbor adapter", "adapt this benchmark for harbor", "harbor adapter for X", "port this benchmark to harbor", or wants to integrate any benchmark dataset into the Harbor evaluation framework (https://harborframework.com).

Use when the user wants to merge a branch, resolve conflicts, integrate remote changes, or recover from a push rejection. Covers push rejection triage, fetch, merge strategy selection, conflict resolution, and verification.

This skill should be used when the user asks to "run subagent-driven development", "execute plan with subagents", "dispatch implementer agents", "use sdd workflow", "sadd", or wants to implement a plan using parallel subagents with spec and quality review loops.

Use at end of a session to update HANDOFF.yaml with completed work, new gaps discovered, and current project state. Also use when asked to create a HANDOFF.yaml for a project that doesn't have one yet.

Use when the user wants to clean up merged branches locally and on remote. Lists merged branches, confirms before deleting, and prunes stale remote refs.

Use when needing a structured knowledge graph of a codebase for exploration or search. gkg index a repo, then use gkg search or gkg context to find definitions, usages, and relationships.

Use at the start of a session to orient to outstanding work — scans for HANDOFF.yaml (or HANDOFF.md) files, triages items by priority, and acts according to risk level without asking for approval on P1/P2 work.

This skill should be used when the user asks to "triage", "what needs fixing", "what's broken", "prioritize issues", "run triage", or wants a P0/P1/P2 issue breakdown across one or more repos.

Use when the user runs "/cap" or asks to "commit and push", "cap it", "ship it", "save progress". Scans for secrets, generates a commit message, stages appropriate files, commits, and pushes.

This skill should be used when the user runs "/eod", asks to "run end of day", "do the eod ritual", "wrap up the session", or "end of day". Executes a closing ritual: run parallel handoff across all active repos (24h git window), then update the Obsidian daily note.

Use when you need to dispatch one or more fast, cheap subagents for parallel or sequential work — research, summarize, transform, verify, batch-process, or any task that doesn't require a specialized agent. Invoke via the Agent tool with subagent_type "atelier:minion".

This skill should be used when the user asks to "run gates", "validate rust", "pre-commit check", "run cargo validation", "check before committing", or wants to run the full Rust validation suite before a commit.

This skill should be used when the user asks to "edit workflow", "fix CI", "check cross-compile", "verify binary", "update github actions", "debug CI failure", "verify target triple", or needs help with CI/CD workflow files or cross-compilation.

This skill should be used when the user asks to "safe to commit", "check merge strategy", "commit safely", "is it safe to merge", "should I rebase or merge", or wants to verify the git strategy is safe before committing or merging.

Write cross-project analysis context back into nested HANDOFF docs after a handup survey or any multi-project analysis session. Appends extra entries (discovery/escalation/note) and a log entry to each touched HANDOFF.yaml, syncs to SQLite, and commits. The write-back counterpart to handup.

This skill should be used when the user runs "/atelier:handover", asks to "visualize the handoff", "show handoff diagrams", "generate a handoff report", "explore the handoff", or provides a path to a HANDOFF file. Reads HANDOFF.*.{yaml,json,md} files and produces a prose summary plus Mermaid diagrams for project orientation.

Surveys the current working directory and all nested subdirectories for outstanding work — HANDOFF.yaml files, TODO/FIXME markers, and open items. Read-only orientation view across projects. Use before handon to decide where to focus.

This skill should be used when the user asks to "show hook status", "hook failures", "what hooks ran", "why did my hook fail", "hook overhead", "list active hooks", or wants visibility into Claude Code hook execution and failures.

Use when a user says "onboard me", "how do I set up atelier", "what does this plugin do", "walk me through setup", or invokes /atelier:onboard. Also suggested by handon when neither atelier nor sanctum appear to have been verified in a live session before.

This skill should be used when the user asks to "apply review fixes", "fix sentinel suggestions", "auto-fix review", "apply code review", "batch apply sentinel fixes", or wants to automatically apply suggestion-level fixes from a sentinel code review report.