> Keeps an agent building and validating real output instead of servicing its own bookkeeping. Use for multi-stage pipelines, capability roadmaps, long-running implementation loops, migrations, staged data or build runs, and any run where administrative hashes, locks, receipts, dashboards, certification markers, or progress metadata can block correct work. Use when an agent refuses to advance, rewinds finished stages, or reruns unchanged work because a marker is missing or stale.
npx skills add https://github.com/Vuk97/forward-implementation-first --skill forward-implementation-first
Build working capability and correct output before administrative bookkeeping.
Apply this contract both while building a system and while running it.
This skill applies to:
It applies whether the action runs through a full orchestrator, a focused
command, or a manual stage invocation.
Before each action, classify it as one of:
adapter, runtime path, schema, fixture, or final output.
schema, counts, samples, conservation, consistency, nontruncation, or
measured resources.
receipts, dashboards, certification markers, progress metadata, or
presence-only records.
Choose categories 1 and 2. Skip category 3 unless the user asks for it or the
artifact is itself part of the product. When administrative work blocks a path
without protecting correctness, remove that dependency from the path.
surfaces.
certification marker, dashboard row, or progress record is missing, stale, or
incompatible.
changed. Replay only that producer's dependency cone.
artifact presence as the product.
from execution paths.
changed dependency cone and inspect the output.
cycle when no defect remains.
A stage may move backward or be replayed only when at least one real condition
holds:
or incompatible with the consumer;
Missing or stale administrative metadata is not one of those conditions.
When the orchestrator refuses to run a stage only because of a receipt,
certification marker, progress record, administrative hash, or lock:
Do not refuse an authorized manual run because the full pipeline cannot issue a
receipt. Do not push the user toward another agent or tool to get around your
own bookkeeping.
Use the checks that match the change:
Hashes may identify inputs or revisions, but they never grant correctness,
execution, or roadmap credit.
Bookkeeping is cheap to skip. Evidence is not. Keep these:
signature your format requires, a hash that is part of the output contract.
Those are features, not paperwork.
are operating on. Getting that wrong means correct work on the wrong target.
end-to-end runs, and whatever your domain treats as proof.
is not a check that the path produced the right answer.
the result, and the expectation it was checked against. Its absence blocks
that specific claim. It does not invalidate unrelated earlier stages.
benchmark, long scan, or anything else that saturates CPU, memory, or a port.
focused-test lanes.
backlog exists, keep up to N cheap-model lanes assigned continuously. Refill
a completed lane with the next independent light task without waiting for the
whole wave. Do not manufacture busywork to occupy slots.
acceptance, and conclusions. Parallel lanes prepare, implement, inspect, and
test nonoverlapping support work. They never become competing truth.
completed lane output after local trust-but-verify checks, when its owning
dependency is reached.
the cone.
producer, consumer, adapter, or output instead.
Report implemented behavior and measured output first. List blockers literally.
Keep infrastructure progress separate from evidence about the output. Do not
turn administrative completion into a substitute for working capability.
Take vuk97/forward-implementation-first from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.