Skip to content

Business operating system for Claude Code — 57 skills, 21 agents, smart daemon. Unified inbox (WhatsApp/Email/Slack/Telegram), autonomous PR merge, full-AWS monitoring, revenue (Stripe+RevenueCat), e-commerce (Shopify), marketing (Klaviyo/Meta/GA4), voice (Bland/ElevenLabs), APM (Datadog/NewRelic/OTEL), YOLO mode.

fix(ops-ci): emit only currently-failing workflows; add pre-dispatch staleness check (#196)

Aurora Capital2cabd935ce5c
diff --git a/claude-ops/.claude-plugin/plugin.json b/claude-ops/.claude-plugin/plugin.json
index ef084120..4797c26d 100644
--- a/claude-ops/.claude-plugin/plugin.json
+++ b/claude-ops/.claude-plugin/plugin.json
@@ -1,6 +1,6 @@
{
"name": "ops",
- "version": "2.0.8",
+ "version": "2.0.9",
"description": "Business operations command center for Claude Code — 30 skills, 14 agents, and a smart background daemon that manage communications (WhatsApp/Email/Slack/Telegram), projects, infrastructure (all AWS services), revenue (Stripe + RevenueCat), e-commerce (Shopify), marketing (Klaviyo/Meta/GA4/GSC), monitoring (Datadog/New Relic/OTEL), and voice (Bland AI/ElevenLabs/Whisper). Includes /gtm go-to-market planner, /ops:projects portfolio dashboard, and YOLO mode for autonomous operation with 4 parallel C-suite agents.",
"author": {
"name": "Lifecycle Innovations Limited",
diff --git a/claude-ops/bin/ops-ci b/claude-ops/bin/ops-ci
index 6b1e1b2c..61db7e07 100755
--- a/claude-ops/bin/ops-ci
+++ b/claude-ops/bin/ops-ci
@@ -1,6 +1,20 @@
#!/usr/bin/env bash
-# ops-ci — Recent CI failures across all repos → JSON
-# Checks last 24h of GitHub Actions failures
+# ops-ci — Currently-failing CI workflows across all repos → JSON
+#
+# Returns workflows where the LATEST run on a tracked branch (main/dev/master)
+# is currently failing. Stale failures that have since been fixed by a newer
+# successful run are excluded — this prevents downstream consumers
+# (ops-fires, ops-go) from dispatching fix agents to already-resolved fires
+# and burning tokens.
+#
+# Algorithm per repo:
+# 1. Pull last 30 runs (any branch, any conclusion) — survey workflows in play
+# 2. Extract distinct workflow names from those runs
+# 3. For each workflow × tracked branch (main, dev, master), fetch latest run
+# 4. Emit only entries where conclusion == "failure"
+#
+# Override: OPS_CI_MODE=legacy reverts to "any failure in last 24h" (slower
+# but matches pre-1.7.x behaviour). Default is "current-state".
set -euo pipefail
@@ -16,19 +30,70 @@ if ! command -v gh &>/dev/null; then
fi
REPOS=$(jq -r '[.projects[].repos[]] | unique | .[]' "$REGISTRY")
-SINCE=$(date -v-1d +%Y-%m-%d 2>/dev/null || date -d '1 day ago' +%Y-%m-%d 2>/dev/null || date +%Y-%m-%d)
+MODE="${OPS_CI_MODE:-current}"
+TRACKED_BRANCHES=("main" "dev" "master")
TMPDIR_OPS=$(mktemp -d)
trap 'rm -rf "$TMPDIR_OPS"' EXIT
+# ─── Per-repo scan ──────────────────────────────────────────────────────────
+scan_repo_current() {
+ local repo="$1"
+ local safe_name
+ safe_name=$(echo "$repo" | tr '/' '_')
+
+ # Survey workflows in recent activity (last 30 runs across all branches)
+ local recent_workflows
+ recent_workflows=$(gh run list --repo "$repo" --limit 30 \
+ --json name,workflowName 2>/dev/null \
+ | jq -r '[.[] | (.workflowName // .name)] | unique | .[]' 2>/dev/null || true)
+
+ [ -z "$recent_workflows" ] && return 0
+
+ local failures="[]"
+ while IFS= read -r workflow; do
+ [ -z "$workflow" ] && continue
+ for branch in "${TRACKED_BRANCHES[@]}"; do
+ # Latest run for this workflow + branch
+ local latest
+ latest=$(gh run list --repo "$repo" --workflow "$workflow" --branch "$branch" --limit 1 \
+ --json name,headBranch,createdAt,conclusion,event,workflowName 2>/dev/null || echo '[]')
+
+ # Skip if no run, or latest is success/skipped/cancelled/in_progress
+ local conclusion
+ conclusion=$(echo "$latest" | jq -r '.[0].conclusion // empty' 2>/dev/null)
+ [ "$conclusion" != "failure" ] && continue
+
+ # Append to failures array
+ failures=$(echo "$failures" | jq --argjson new "$latest" '. + $new')
+ done
+ done <<< "$recent_workflows"
+
+ if [ "$(echo "$failures" | jq 'length')" -gt 0 ]; then
+ echo "{\"repo\":\"$repo\",\"failures\":$failures}" > "$TMPDIR_OPS/$safe_name.json"
+ fi
+}
+
+scan_repo_legacy() {
+ local repo="$1"
+ local safe_name
+ safe_name=$(echo "$repo" | tr '/' '_')
+
+ local failures
+ failures=$(gh run list --repo "$repo" --status failure --limit 3 \
+ --json name,headBranch,createdAt,conclusion,event 2>/dev/null || echo '[]')
+
+ if [ "$failures" != "[]" ] && [ -n "$failures" ]; then
+ echo "{\"repo\":\"$repo\",\"failures\":$failures}" > "$TMPDIR_OPS/$safe_name.json"
+ fi
+}
+
for repo in $REPOS; do
(
- SAFE_NAME=$(echo "$repo" | tr '/' '_')
- FAILURES=$(gh run list --repo "$repo" --status failure --limit 3 \
- --json name,headBranch,createdAt,conclusion,event 2>/dev/null || echo '[]')
-
- if [ "$FAILURES" != "[]" ] && [ -n "$FAILURES" ]; then
- echo "{\"repo\":\"$repo\",\"failures\":$FAILURES}" > "$TMPDIR_OPS/$SAFE_NAME.json"
+ if [ "$MODE" = "legacy" ]; then
+ scan_repo_legacy "$repo"
+ else
+ scan_repo_current "$repo"
fi
) &
done
diff --git a/claude-ops/skills/ops-fires/SKILL.md b/claude-ops/skills/ops-fires/SKILL.md
index efc07f21..044bbaa6 100644
--- a/claude-ops/skills/ops-fires/SKILL.md
+++ b/claude-ops/skills/ops-fires/SKILL.md
@@ -185,6 +185,29 @@ If no fires: show "ALL SYSTEMS OPERATIONAL" with last-checked timestamps.
---
+## Pre-dispatch staleness check (MANDATORY)
+
+The pre-gathered CI data is cached and may be minutes-to-hours old. Before
+dispatching ANY fix agent, verify the failure is still red on its branch HEAD.
+This is defense-in-depth: even after the `bin/ops-ci` "current-state" filter
+(which only emits workflows whose latest run on a tracked branch is failing),
+a fix may have landed in the seconds since the cache was written. Dispatching
+to a self-resolved fire wastes Sonnet quota — typically 50–150k tokens per
+agent before it figures out there's nothing to fix.
+
+For each fire the user selects:
+
+```bash
+gh run list --repo "$REPO" --workflow "$WORKFLOW" --branch "$BRANCH" --limit 1 \
+ --json conclusion,databaseId,createdAt --jq '.[0]'
+```
+
+- If `conclusion == "success"` → SKIP. Mark task completed with metadata `{resolution: "self-resolved-pre-dispatch"}`. Do NOT spawn agent.
+- If `conclusion == "failure"` → proceed to dispatch.
+- If `conclusion == null` (in_progress) → wait 30s, recheck once, then proceed if still null.
+
+For workflows scoped only to PRs (no main/dev runs), check the PR's combined CI status instead: `gh pr checks <num> --repo "$REPO" --json bucket,name`.
+
## Dispatch fix agent
When user selects to fix an issue, use `AskUserQuestion` to confirm the scope before dispatching: