actual-software

Harvest walkthrough snapshot output and update curriculum READMEs to match what walkthroughs actually produce. Walkthroughs save their output to test-harness/walkthrough-snapshots/<lesson>/. This skill reads those snapshots, compares against README claims, and updates the READMEs so students see accurate commands and output. Use whenever editing curriculum READMEs, walkthrough scripts, or lesson packs. Also trigger on "validate lessons", "update lesson content", "sync READMEs with walkthroughs", "curriculum drift", or any question about whether what students are told to do matches reality.

Clean up and run this repo's tutorial walkthrough harness safely. Use before or after `test-harness/tutorial-walkthrough.sh`, `test-harness/behavioral-smoke.sh`, or `$validate-lesson-content`; when `sfi-walkthrough-*` cities, scratch Dolt/tmux/agent processes, or hook-owned walkthrough runs interfere; or when validating lesson snapshots in `test-harness/walkthrough-snapshots`.

Harvest walkthrough snapshot output and update curriculum READMEs to match what walkthroughs actually produce. Walkthroughs save their output to test-harness/walkthrough-snapshots/<lesson>/. This skill reads those snapshots, compares against README claims, and updates the READMEs so students see accurate commands and output. Use whenever editing curriculum READMEs, walkthrough scripts, or lesson packs. Also trigger on "validate lessons", "update lesson content", "sync READMEs with walkthroughs", "curriculum drift", or any question about whether what students are told to do matches reality.

Feature-complete companion for the actual CLI, an ADR-powered CLAUDE.md/AGENTS.md generator. Runs and troubleshoots actual adr-bot, status, auth, config, runners, and models, and drives the Actual Advisor (login, advisor, whoami, logout) for org-scoped architecture questions. Covers all 5 runners (claude-cli, anthropic-api, openai-api, codex-cli, cursor-cli), all model patterns, all 3 output formats (claude-md, agents-md, cursor-rules), and all error types. Use when working with the actual CLI, running actual adr-bot, configuring runners or models, troubleshooting errors, managing output files, or asking the Advisor what your team's ADRs say about a design choice.

Feature-complete companion for the actual CLI, an ADR-powered CLAUDE.md/AGENTS.md generator. Runs and troubleshoots actual adr-bot, status, auth, config, runners, and models, plus the Actual AI platform surface: login, logout, whoami, and advisor (org-scoped architecture Q&A). Covers all 5 runners (claude-cli, anthropic-api, openai-api, codex-cli, cursor-cli), all model patterns, all 3 output formats (claude-md, agents-md, cursor-rules), and all error types. Use when working with the actual CLI, running actual adr-bot, signing in to Actual AI, asking the advisor, configuring runners or models, troubleshooting errors, or managing output files.