
Claude Skills by tomcounsell
github.com/tomcounsellUse when onboarding to the Valor AI codebase, understanding the system architecture, or when the user asks 'how does this work'. Comprehensive codebase orientation guide.
Use when reading SMS or iMessage from the macOS Messages app. Triggered by requests to get 2FA codes, check recent texts, or search message history.
Single-stage router for development work. Assesses current state, dispatches ONE sub-skill, then returns. The PM session handles pipeline progression.
Check Sentry for unresolved issues and run triage on demand. Triggered by requests to check Sentry errors, run Sentry triage, or handle Sentry issues.
Use when configuring a new machine to run the Valor Telegram bridge. Triggered by 'setup', 'configure this machine', or 'new machine setup'.
Use when reading or sending Telegram messages. Triggered by requests to check recent messages, search conversation history, or send messages/media to chats.
Use when deploying updates to this machine: pull, sync deps, verify, restart services. Triggered by 'update', 'deploy', 'pull and restart', or after git pull.
Use when browsing x.com (Twitter), reading the timeline, posting, replying, liking, or checking DMs. Triggered by requests to post on X, browse the X feed, comment on tweets, or read/reply to X DMs.
Re-sync this repo's Claude Code config (.claude/agents, .claude/commands, settings.json hooks/permissions) into OpenCode's .opencode/ layout, idempotently. Use after any change to .claude/agents/, .claude/commands/, .claude/settings.json, or .claude/hooks/.
Claude Code hooks have grown organically to 8 hook commands across 5 event types, with inconsistent safety practices. A manual audit uncovered: silent error swallowing (`|| true` without logging), `set -e` in bash hooks (fatal in hooks), missing `log_hook_error()` calls, and no codified best practices. When new hooks are added, there is no automated check that they follow established patterns.
A targeted repair skill for fixing test failures and PR review blockers. Invoked by the user directly or automatically by `/do-build` at lifecycle failure points.