hcaiano
UserHenrique Caiano's custom agent skills (Claude, Codex, herdr). Install with: npx skills@latest add hcaiano/skills
Categories
Indexed Skills (12)
check-logs
Check logs in an existing turborepo dev TUI inside herdr. Use when the user asks about an app's dev output, whether an app is running, or an error shown by a `bun dev` / `turbo run dev` pane. Read-only: navigate the existing TUI; never start, stop, or restart the server.
goal-loop
Manual-only long improvement campaign run by Claude + Codex as a pair. Use only when the user explicitly invokes goal-loop and supplies one target plus what 'better' means; the agents repeatedly test, improve, ship PRs, and require current evidence plus peer acceptance to stop.
no-slop
Draft or edit Henrique's writing in his voice, or flag concrete AI-slop patterns without rewriting. Use for English or European Portuguese.
ask-peer
Ask the opposite model for one focused opinion, review, or scoped pass without a live Herdr pair. Codex may consult Fable autonomously; Claude Code consults Codex only when explicitly invoked.
review-pr-comments
Manual-only cleanup of actual feedback on an existing PR: verify comments, batch valid in-scope fixes, respond, and leave the PR merge-ready.
art-director
Manual-only visual exploration before implementation. Invoke explicitly to explore and choose interface, product, campaign, or brand directions using imagegen, optional brandkit proof, and one-question-at-a-time intake.
herdr-orchestrate
Manual-only orchestration of GitHub issues in dedicated Herdr work units: triage, delegate, scope-check, ship, merge, and dismantle. The user owns blocked decisions and holds for high-risk or visible UI work.
cmux-pair
Legacy cmux-based pair programming bootstrap for Claude and Codex talking through cmux surfaces with an XML protocol. Keep this for historical reference and migration only. Prefer herdr-pair for current Claude/Codex collaboration; use this skill only when the user explicitly asks for the old cmux workflow.
cyber-audit
Read-only audit of this machine against a named CVE, malicious package, or supply-chain advisory. Use only when explicitly asked whether this machine is affected; not for general security review, news, breach response, or remediation.
debug-mode
Evidence-driven debugging for unresolved, flaky, environment-specific, or production-only bugs. Use when asked to diagnose or when normal inspection cannot prove the cause; skip obvious errors and already-proven causes.
ship-it
Ship finished work: the graded local review gate, then one PR carried to green CI. Use when the user wants to ship, open, or update a PR, or when another skill needs that graded gate.
herdr-pair
Persistent Claude-Codex pairing inside one Herdr tab. Use for live peer work, workflows requesting a pair, or any inbound `[agent ...]` / `[herdr-pair control ...]` line — including after context compaction; resume the exact session and reply through the bundled transport.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.