Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Use when testing or verifying an Umbrel App Store package, including linting, Umbrel device or local umbrelOS test environments, fresh install and update-path testing, browser verification, logs, persistence, dependencies, and PR evidence.
BIND 9's memory allocator wrapper (isc_mem memory contexts and isc_mempool fixed-size pools). Use when writing or reviewing code that allocates/frees memory anywhere in BIND 9, when choosing between isc_mem_get/put and isc_mem_allocate/free, when debugging "isc_mem_inuse(ctx) == 0" or mempool-leak assertion failures, memory-leak reports, overmem/water-mark behavior, or ISC_MEM_DEBUG* tracing.
Methodology for root-causing hard concurrency / memory-ordering bugs (intermittent races, use-after-free, RCU/lock-free publish-order defects, "impossible" stale reads) with LTTng flight-recorder (snapshot) tracing — when static analysis, printf, and a debugger all fall short. Covers the snapshot+violation+abort setup, tracepoint instrumentation discipline (why tracepoints not printf), how to enrich a violation event so the trace is self-diagnosing, and the trace-reading patterns that crack these bugs (notably: a stale-after-write "paradox" is a happens-before gap, not a timing bug).
Working on Holmake itself — modifying `tools/Holmake/`, diagnosing why a target rebuilds spuriously, debugging the dep graph, understanding the three-phase `bin/build` flow, the heap-as-implicit-dep mechanism, BIC node semantics, the `HFS_NameMunge` `.hol/objs/` indirection, `--rebuild=mtime` vs `--rebuild=cachekey`, mosml/poly portability of `HM_Cline` options, and how to write Holmake regression tests. Use this skill whenever the task touches `tools/Holmake/**`, `tools-poly/build.sml`, `tools/sequences/*`, Holmakefile syntax, build-order diagnostics, or "why is this rebuilding?" investigations.
Run the full local verification gate for lerd before committing — build, test, vet, gofmt, UI tests, installer tests, then install the local build and smoke-test. Use before every commit or PR, and whenever asked to "verify", "check CI locally", or "make sure it's green".
Validate that two tables or query results are identical — or diagnose exactly how they differ. Discover schema, identify keys, profile cheaply, then diff. Use for migration validation, ETL regression, and query refactor verification.
Cloudflare-style AI code review for dbt/SQL pull requests. Produces a signed APPROVE/COMMENT/REQUEST_CHANGES verdict where every blocking finding is backed by a deterministic engine call — column-lineage blast radius, query equivalence, PII classification, and A–F grade. Use to review a dbt PR or the working-tree changes before merge.
REQUIRED after building or modifying ANY dbt model that has columns declared in `schema.yml` / `_models.yml`. Run `altimate-dbt schema-verify --model <name>` to diff actual columns against the spec, and treat any `mismatch` verdict as "not done." The most common reason "the build is green but the tests still fail" is that the model produces the right *data values* in the wrong *column shape* — extra columns, missing columns, wrong order, wrong types. Many dbt equality tests grade the column tuple `(name, type, position)` exactly, and the agent's prior bias is to add "helpful" extras (`p1`/`p2`/`p3` rank breakdowns, name-resolved variants, lineage metadata) or reorder columns "more logically." Both break the contract. This skill enforces the mechanical check that catches those bugs before declaring done. Use it before declaring any model task complete.
Debug dbt errors — compilation failures, runtime database errors, test failures, wrong data, and performance issues. Use when something is broken, producing wrong results, or failing to build. Powered by altimate-dbt.