All skills
38,472 Skills
Search uses this published catalog; rankings and counts are snapshots.
codebase-knowledge-builder
Study unfamiliar codebases and produce evidence-backed knowledge artifacts. Use for repository orientation, architecture mapping, subsystem tracing, onboarding, or codebase documentation.
instagram-marketing
Plan, draft, audit, and publish content for Instagram. Use when the user wants to write a caption with a first-125-char hook, plan a slide-by-slide carousel, reverse-engineer the hook from a viral Reel or carousel, size hashtags the 2026 way, remove AI tells from a caption, or pl
audit-agent-efficiency
Read a finished agent's transcript, find the turns it wasted, and put a fix on disk for each one (a CLAUDE.md line, a script, a breadcrumb, a pattern) so the next run is fast. Triggers on: 'audit that agent', 'why was that slow', 'review the agent's run', or /audit-agent-efficien
build-debug-tool
Turn a repeated hunt through logs, JSON dumps or database rows into a permanent diagnostic tool, at the moment you notice the pain. Triggers on: 'build a debug tool', 'make this dump readable', 'I keep grepping for this', 'I keep typing this python -c', or /build-debug-tool. Test
cross-repo-skill-comparison
Compare the Claude skills across two or more repos (paths or GitHub links) and write one HTML page: which skills exist where, which copy is newest, what each fork changed, which are worth pulling into this repo. Triggers on: 'compare skills across repos', 'which repo has the late
debuggability-audit
Find where a failure leaves no usable evidence, in the logs or on the screen: swallowed exceptions, context-free logs, silent fallbacks, errors caught and only logged, error fields no component reads; scored out of 10, guard mode checks the diff. Triggers on: 'is this debuggable'
handoff
Compact the current conversation into a handoff document a fresh agent can resume from: where the work stands, what was decided and why, what is next, which files to open first. User-invoked only: /handoff.
hk-arch-audit
Audit a codebase against eight agent-native principles and write one page in writeup/ with a score out of 10. Triggers on: /hk-arch-audit.
hk-ask-until-clarity
Before starting a task that could be read two ways, ask the person numbered questions, each with a recommended answer, until the goal, what done looks like and what is out of scope are settled; then write the agreed brief down. Triggers on: 'ask until clear', 'make sure you under
hk-autonomy-audit
Judge whether an agent could take one area from symptom to verified fix with no person in the loop, and write up the gaps as one committed markdown page in docs/loop_audits/ with a score out of 10. Triggers on: 'loop audit', 'agent readiness', 'could an agent fix this alone', 'ca
hk-breadcrumb-creator
Trace a flow through the code and write its breadcrumb (FLOW.md, DETAILS.md, DEBUG.md) under docs/breadcrumb_analysis/, with the index row. Triggers on: 'write a breadcrumb', 'add a breadcrumb', 'trace this flow', 'document this flow', 'write a debug guide for', or /hk-breadcrumb
hk-changelog
Write a CHANGELOG.md entry from the branch's diff against main, then commit it on a changelog branch and open a PR. Triggers on: 'update the changelog', 'changelog entry', 'add to the changelog', or /hk-changelog.
hk-chart-validation
Check that the charts in a codebase tell the truth: chart type, axis baselines, aggregation, units, labels, tooltips and how they render in Chrome at each width; writes the findings as one HTML page with a score out of 10. Triggers on: 'validate the charts', 'is this chart mislea
Catalog captured:2026-11-21