watchdog
SolidGeneric watchdog for stuck or stale automation. Inspects one or more action classes, determines whether they are merely slow vs actually stuck/stale, and takes the smallest safe completion action. Includes the autopilot draft-PR watchdog: stale draft PRs should be completed and marked ready for review, not merely surfaced as merge nudges.
Install
Quality Score: 83/100
Skill Content
Details
- Author
- mifunedev
- Repository
- mifunedev/openharness
- Created
- 5 months ago
- Last Updated
- today
- Language
- TypeScript
- License
- Apache-2.0
Integrates with
Similar Skills
Semantically similar based on skill content — not just same category
autopilot
Self-improvement loop: each hourly run implements the oldest open issue labeled `autopilot` that has no open PR. When the queue is empty it runs first-principles `/audit harness` research, files its own `autopilot` ticket from the top-ranked finding, and builds that. Decomposes via the pm agent, then runs `/ship-spec --issue`, which owns the build end-to-end (the two compacts bracketing implement, a worktree Advisor, `/delegate` workers running `.oh/scripts/ralph.sh`, the `/eval` gate, and the `/audit pr` promotable undraft) through to a ready-for-review PR whose description states why this item was selected. Harness-infra only (skills/rules/docs/scripts/crons/wiki) — never sandbox application code. Runs in its own per-run tmux session. Caps: 6 open autopilot PRs created per UTC day AND 10 total open; never auto-merges. TRIGGER when: the hourly .oh/crons/autopilot.md fires, or invoked manually on demand (e.g. /autopilot --dry-run to preview the next selection).
wise-pr-watch-auto
Autonomous variant of `/wise-pr-watch` — drive the current branch's PR to merge in bulk rounds with NO prompts. Each round: one linear 2-minute poll until CI is terminal and every review bot that is going to review the head (Copilot, CodeRabbit) has done so; gather every failing check, every unresolved bot thread (outdated ones included), every open Sonar issue; fix the whole set in one pass, resolve every handled thread immediately, one commit, one push; wait one 2-minute re-review window to see what the push triggered (the push IS the trigger — no `@coderabbitai review` while a bot is auto-reviewing); settle again. Converges on the first head with nothing actionable, or after two nit-only rounds (remaining nits accepted and resolved, no push), or at the round cap. Reads the base branch's rules up front (thread-resolution rule, required approvals), re-reads the PR state at every tick so a PR merged or closed from outside ends the run, and keeps its state under the PR so a re-invocation resumes. A stuck bot g
goalforge-watchdog
goalforge-internal — use entry commands; do not auto-trigger Semantic spec-vs-diff gap audit for an SDD work package. Invoked by goalforge-verify AFTER acceptance criteria pass: reconstructs the WP goal/contract, reads the changed files plus their neighbors, and reports gaps — claimed-vs-implemented mismatches, missing tests/docs at changed seams, and deviations from the spec constraints/boundaries. Light summary by default (into findings.md; material gaps recorded as recap.md loop-backs via goalforge-recap); deep verify-gap.md report on opt-in. Advisory only — never blocks or rewrites goalforge-verify's status-advance authority. Trigger: goalforge-verify delegates a gap audit for a WP, or a user asks to audit a WP diff against its spec.