donaldfilimon/abi / .claude/skills/complete-base
complete-base
VerifiedBuild the abi CLI and drive the base completion path — `abi complete "<input>"` with no flags — routing to the local model, running the constitution audit, and optionally recording the completion in WDBX. Use to smoke-test the core local-completion path after touching crates/abi-ai or the model catalog. Fully local: no --live (remote) and no --learn (SEA).
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.3 KB |
| complete.shexecutable | script | 1.8 KB |
complete.sh
View on GitHub#!/usr/bin/env bash
# complete-base driver: build the abi CLI and drive the BASE completion path —
# `abi complete "<input>"` with no flags. This routes to the local model
# (claude-fable-5 by default), runs the constitution audit, and records the
# completion in the default WDBX store. It is fully local: no `--live` (no remote
# provider) and no `--learn` (that SEA path is covered by sea-learn-loop).
# Asserts exit codes + output markers. Resolves repo root from own path.
#
# Usage: .claude/skills/complete-base/complete.sh ["prompt"] [model-id]
set -uo pipefail
SCRIPT_DIR=$(cd -- "$(dirname -- "${BASH_SOURCE[0]}")" && pwd)
REPO_ROOT=$(cd -- "$SCRIPT_DIR/../../.." && pwd)
cd "$REPO_ROOT"
ABI="$REPO_ROOT/target/debug/abi"
PROMPT="${1:-complete-base probe: summarize the scheduler status}"
MODEL="${2:-}"
fail=0
say() { printf '\n=== %s ===\n' "$*"; }
markers() { local out="$1"; shift; for m in "$@"; do
grep -qF -- "$m" <<<"$out" && echo "[ok] marker: $m" \
|| { echo "[FAIL] missing marker: $m"; fail=$((fail+1)); }; done; }
say "build cli"
./tools/cargo.sh build -p abi-cli >/dev/null 2>&1 && echo "[ok] build" || { echo "[FAIL] build"; exit 1; }
[ -x "$ABI" ] || { echo "[FAIL] $ABI not produced"; exit 1; }
if [ -n "$MODEL" ]; then
say "abi complete --model $MODEL"
out=$("$ABI" complete --model "$MODEL" "$PROMPT" 2>&1); rc=$?
else
say "abi complete (default model)"
out=$("$ABI" complete "$PROMPT" 2>&1); rc=$?
fi
printf '%s\n' "$out"
[ "$rc" -eq 0 ] || { echo "[FAIL] complete exit $rc"; fail=$((fail+1)); }
markers "$out" "model=" "audit_passed=true" "persisted=true" "wdbx kv_entries="
say "summary"; echo "failed checks: $fail"
[ "$fail" -eq 0 ] && echo "RESULT: PASS — base completion ran local + persisted." || echo "RESULT: FAIL — $fail check(s)."
exit "$fail"
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |