Skip to content

Automation bots and maintenance control loop

This page is the operational index for the repo's maintenance bots, especially the GitHub Advanced Security (GHAS) and upgrade-review automations that keep the project moving without waiting for a manual audit.

Why these bots exist

The repo already exposes strong local/operator commands such as:

  • python -m sdetkit intelligence upgrade-audit --format json --top 10
  • python -m sdetkit kits optimize --goal "upgrade umbrella architecture with agentos optimization" --format json
  • python -m sdetkit kits expand --goal "add more helpful automation bots and GitHub Advanced Security coverage" --format json

The bots turn those same maintenance signals into a recurring GitHub-native operating loop:

  1. collect the state,
  2. summarize the risk,
  3. open or refresh an actionable issue,
  4. point maintainers to the next safe validation lane.

Current bot inventory

Worker templates that align with the bot surface

In addition to scheduled GitHub bots, the repo now ships AgentOS worker templates that let maintainers run the same expansion/review loops on demand and keep the outputs deterministic:

  • repo-expansion-control β€” runs kits optimize + kits expand, writes JSON artifacts, and bundles them for roadmap / dashboard follow-up.
  • adapter-smoke-worker β€” validates optional notification adapters, captures integration-adapter route maps, and packages a quickstart-ready bundle.
  • dependency-radar-worker β€” captures an offline-safe upgrade inventory, radar snapshot, and validation route map for refactor / dependency review.
  • validation-route-worker β€” turns a package or domain query into a route map plus doctor-linked upgrade guidance.
  • runtime-watchlist-worker β€” emits a runtime-core fast-follow watchlist plus a route map for the hottest runtime validation lanes.
  • docs-search-radar β€” runs a strict MkDocs build, captures the build log, and bundles docs-search evidence for later review.
  • integration-topology-worker β€” validates heterogeneous topology proof, snapshots optimize guidance, and bundles architecture-ready evidence for premium-gate follow-up.
  • release-readiness-worker β€” snapshots doctor plus github_automation_check so release-readiness follow-up can happen outside of a publish crunch.
  • worker-alignment-radar β€” keeps expansion recommendations, automation coverage, and template inventory aligned with the repo's active worker surface.

Recommended local launches:

python -m sdetkit agent templates run repo-expansion-control
python -m sdetkit agent templates run adapter-smoke-worker
python -m sdetkit agent templates run dependency-radar-worker
python -m sdetkit agent templates run validation-route-worker --set query=httpx
python -m sdetkit agent templates run runtime-watchlist-worker
python -m sdetkit agent templates run docs-search-radar
python -m sdetkit agent templates run integration-topology-worker
python -m sdetkit agent templates run release-readiness-worker
python -m sdetkit agent templates run worker-alignment-radar
python -m sdetkit kits expand --goal "add more bots workers search and repo expansion" --format json

Security and GitHub Advanced Security bots

  • security.yml β€” runs CodeQL.
  • osv-scanner.yml β€” runs OSV vulnerability scanning and uploads SARIF.
  • dependency-audit.yml β€” runs pip-audit against the repo dependency surface.
  • sbom.yml β€” refreshes SBOM coverage for downstream review.
  • security-maintenance-bot.yml β€” creates the weekly maintenance checklist and weak-spot report.
  • secret-protection-review-bot.yml β€” creates a weekly secret-protection review issue for push protection, delegated bypass, validity checks, and open secret-scanning backlog slices.
  • ghas-review-bot.yml β€” creates the weekly GHAS digest issue with open-alert counts, workflow freshness, and follow-up prompts.
  • ghas-campaign-bot.yml β€” creates a weekly GHAS campaign planner issue for Copilot Autofix-aware code scanning backlog slices, secret scanning age buckets, and push-protection follow-up.
  • ghas-alert-sla-bot.yml β€” creates a weekly GHAS SLA tracker issue that forces 7/14/30- backlog slices into an owned remediation lane.
  • ghas-metrics-export-bot.yml β€” exports a weekly GHAS metrics artifact plus an issue-driven snapshot of alert totals, age buckets, and workflow freshness.
  • ghas-codeql-hotspots-bot.yml β€” creates a weekly CodeQL hotspot issue that groups the code-scanning queue by rule and file so fixes can land in batches.
  • security-configuration-audit-bot.yml β€” creates a monthly GHAS configuration audit issue covering repo-local workflow coverage, code security configuration visibility, and dependency submission posture.
  • dependency-review.yml β€” blocks pull requests that introduce vulnerable dependencies or denied license drift before they merge.

Dependency and repo-maintenance bots

  • dependency-auto-merge.yml β€” auto-merges safe Dependabot updates after checks pass.
  • pre-commit-autoupdate.yml β€” refreshes pre-commit hook pins.
  • dependency-radar-bot.yml β€” publishes a recurring dependency radar issue plus a runtime fast-follow watchlist.
  • adapter-smoke-bot.yml β€” publishes a weekly adapter smoke issue and artifact bundle for notification adapter quality and expansion readiness.
  • repo-optimization-bot.yml β€” publishes a weekly repo-optimization control loop issue from kits optimize, kits expand, and the GitHub automation maintenance check.
  • workflow-governance-bot.yml β€” publishes a monthly workflow-governance audit for permissions, SHA pinning, and manual recovery posture.
  • docs-experience-bot.yml β€” publishes a weekly docs-experience radar issue plus JSON artifact covering navigation coverage, search posture, and flagship-doc visibility.
  • integration-topology-radar-bot.yml β€” publishes a weekly integration-topology radar issue and worker artifacts so topology proof stays aligned with premium-gate-ready refactors.
  • release-readiness-radar-bot.yml β€” publishes a weekly release-readiness radar issue with doctor output, release asset freshness, and release-workflow coverage.
  • runtime-watchlist-bot.yml β€” publishes a weekly runtime fast-follow issue and worker artifacts so hot-path runtime upgrades stay reviewable.
  • worker-alignment-bot.yml β€” runs the aligned worker templates together and publishes a weekly worker/radar issue with deterministic artifact links, including topology coverage.
  • contributor-onboarding-bot.yml β€” keeps contributor guidance discoverable.
  • pr-helper-bot.yml / pr-quality-comment.yml β€” keep pull requests reviewable and actionable.

New additions in this upgrade

ghas-review-bot.yml

This bot gives the repo a GHAS-native triage layer instead of relying on maintainers to click through the Security tab manually.

It now:

  • counts open code scanning, Dependabot, and secret scanning alerts when the API is available,
  • records workflow freshness for the main security automations,
  • opens a date-scoped πŸ›‘οΈ GHAS weekly digest (...) issue,
  • closes older GHAS digest issues automatically,
  • degrades safely when an additional token is required for private security endpoints.

Optional hardening: set GHAS_READ_TOKEN if you want the bot to use a token with broader security-read access than the default GITHUB_TOKEN.

ghas-campaign-bot.yml

This bot turns the raw GHAS alert queue into a remediation-planning surface instead of a passive dashboard.

It now:

  • groups open code scanning alerts by severity and age,
  • highlights CodeQL/Copilot Autofix candidate volume,
  • tracks secret-scanning age buckets and push-protection bypass follow-up,
  • records repository security-configuration context when the API exposes it,
  • opens a date-scoped 🧭 GHAS campaign planner (...) issue.

security-configuration-audit-bot.yml

This bot gives the repo a monthly control loop for newer GHAS operating-model features that are easy to forget once the initial setup is done.

It now:

  • audits whether the expected GHAS workflows are present in the repo,
  • verifies local supporting config files such as dependabot.yml and codeql-config.yml,
  • queries the repository's attached code security configuration when available,
  • snapshots open alert counts across code scanning, secret scanning, and Dependabot,
  • opens a date-scoped 🧱 GHAS configuration audit (...) issue.

ghas-alert-sla-bot.yml

This bot adds an explicit age-based operating contract to the GHAS queue so stale findings cannot sit in the Security tab indefinitely without an owner.

It now:

  • counts 7+, 14+, and 30+ backlog slices across code scanning, Dependabot, and secret scanning,
  • highlights 14+ high-severity code/dependency findings,
  • tracks push-protection bypass secret alerts that should be treated as same-week work,
  • groups older CodeQL findings by rule so maintainers can batch-fix them instead of triaging one-by-one,
  • opens a date-scoped ⏱️ GHAS alert SLA tracker (...) issue.

secret-protection-review-bot.yml

This bot adds a focused control loop for the newer GHAS secret-protection settings that are easy to enable once and then forget.

It now:

  • reviews the open secret scanning backlog together with push-protection bypass follow-up volume,
  • audits repository-level secret-protection settings such as push protection, delegated bypass, delegated alert dismissal, validity checks, generic secrets, and non-provider patterns,
  • highlights the most common open secret types for faster triage,
  • opens a date-scoped πŸ”‘ Secret protection review (...) issue.

ghas-metrics-export-bot.yml

This bot turns GHAS operations into reusable evidence instead of ephemeral issue text.

It now:

  • exports build/ghas-metrics.json as a workflow artifact,
  • captures alert totals, age buckets, and severity distributions for code scanning, Dependabot, and secret scanning,
  • records workflow freshness for the repo's GHAS automation surface,
  • opens a date-scoped πŸ“Š GHAS metrics snapshot (...) issue,
  • gives dashboards, weekly reviews, and future evidence packs a machine-readable GHAS baseline.

ghas-codeql-hotspots-bot.yml

This bot gives the repo a CodeQL-first remediation lane instead of treating the code-scanning queue as a flat list.

It now:

  • groups open code scanning alerts by rule, tool, and file/path,
  • highlights 14+ and 30+ hotspots so older findings are easier to batch-fix,
  • exports build/ghas-codeql-hotspots.json for follow-up automation or dashboard reuse,
  • opens a date-scoped πŸ§ͺ GHAS CodeQL hotspots (...) issue.

repo-optimization-bot.yml

This bot turns the repo's own umbrella-intelligence surfaces into a recurring enhancement and refactor control loop.

It now:

  • runs sdetkit kits optimize and sdetkit kits expand with the repo's automation/GHAS goal,
  • snapshots the github_automation_check maintenance signal beside those results,
  • publishes a single issue with top feature candidates, GHAS update tracks, search missions, rollout tracks, and innovation opportunities,
  • uploads reusable JSON artifacts for future roadmap, dashboard, or AgentOS consumption.

workflow-governance-bot.yml

This bot adds a repo-wide automation hardening loop beyond GHAS-specific alert triage.

It now:

  • audits every workflow for a top-level permissions block,
  • flags reusable actions that are not pinned to a full commit SHA,
  • flags scheduled workflows that are missing workflow_dispatch manual recovery,
  • exports build/workflow-governance-audit.json for deterministic review,
  • opens a date-scoped 🧾 Workflow governance audit (...) issue.

dependency-radar-bot.yml

This bot turns the repo's existing upgrade-audit intelligence into a scheduled maintenance surface.

It now:

  • generates build/dependency-radar.json,
  • creates a markdown summary with the hottest packages and validation lanes,
  • builds a runtime fast-follow watchlist for hot-path runtime dependencies,
  • uploads those artifacts to Actions,
  • opens a date-scoped πŸ“‘ Dependency radar + runtime watchlist (...) issue.

docs-experience-bot.yml

This bot turns the docs site into a first-class maintenance lane instead of something maintainers only notice after navigation drift has accumulated.

It now:

  • audits MkDocs navigation coverage against the actual docs/ tree,
  • highlights missing nav targets and high-value orphan docs that are outside the primary user journeys,
  • checks that flagship entrypoint pages stay both present and visible in the main nav,
  • records docs-search posture such as search suggest/highlight/share and instant navigation support,
  • opens a date-scoped πŸ“š Docs experience radar (...) issue,
  • exports build/docs-experience-radar.json for dashboards, docs cleanup planning, or later AgentOS consumption.

release-readiness-radar-bot.yml

This bot gives the repo a recurring release-ops review lane so publishing posture is not reconstructed manually right before a release.

It now:

  • runs python -m sdetkit doctor --format json and captures the result as part of the release snapshot,
  • reviews the presence of core release workflows such as release.yml, pages.yml, versioning.yml, and docs-link-check.yml,
  • audits key release trust assets including CHANGELOG.md, ROADMAP.md, RELEASE.md, and the public release docs,
  • reuses the GitHub automation maintenance check to highlight any required workflow/config gaps,
  • opens a date-scoped πŸš€ Release readiness radar (...) issue,
  • uploads build/release-readiness-radar.json plus the raw doctor/maintenance payloads as workflow artifacts.

integration-topology-radar-bot.yml

This bot turns the repo's topology-aware integration proof into a recurring architecture review lane instead of something maintainers only inspect during large refactors.

It now:

  • runs integration-topology-worker on a weekly cadence,
  • captures the bundled topology-check and optimize payloads as reusable worker artifacts,
  • opens a date-scoped πŸ•ΈοΈ Integration topology radar (...) issue,
  • keeps heterogeneous service proof aligned with premium-gate-ready umbrella changes.

worker-alignment-bot.yml

This bot turns the worker layer into a first-class operating lane instead of leaving the templates as individual commands maintainers have to remember.

It now:

  • runs dependency-radar-worker, docs-search-radar, integration-topology-worker, release-readiness-worker, and worker-alignment-radar on a weekly cadence,
  • uploads both the per-worker run records and the template output directories as reusable artifacts,
  • opens a date-scoped πŸ€– Worker alignment radar (...) issue,
  • keeps the worker surface aligned with dependency, docs, release, and automation-review loops.
  1. Open the latest GHAS digest issue.
  2. Open the latest GHAS campaign planner issue and group any 14+ backlog into a focused remediation lane.
  3. Open the latest GHAS alert SLA tracker issue and convert every 14+ breach into an owned issue or PR.
  4. Review the GHAS metrics snapshot artifact before reporting progress or trend direction.
  5. Review the GHAS CodeQL hotspots issue and batch the top rule or top file into one remediation lane.
  6. Review the secret protection review issue and clear any push-protection bypass or delegated-bypass follow-up.
  7. Clear or triage high-severity security findings first.
  8. Open the latest dependency radar issue.
  9. Open the latest repo optimization control loop issue and pick one now candidate or search mission.
  10. Open the latest docs experience radar issue and fold the highest-value orphan docs or missing flagship links back into the canonical journeys.
  11. Open the latest integration topology radar issue before topology, platform, or premium-gate changes expand.
  12. Open the latest release readiness radar issue before a publish window or docs release push.
  13. Open the latest worker alignment radar issue and refresh any stale worker/template lanes before they drift from the base automation surface.
  14. Review the monthly workflow governance audit issue and close any pinning/permissions drift.
  15. Record implementation follow-up in docs/roadmap.md and the weekly maintenance issue.

Optional local dry runs

You can dry-run the same intelligence locally before relying on the bots:

python -m sdetkit intelligence upgrade-audit --format json --used-in-repo-only --top 10
python -m sdetkit intelligence upgrade-audit --impact-area runtime-core --repo-usage-tier hot-path --format md
python -m sdetkit kits optimize --goal "add more helpful automation bots and GitHub Advanced Security coverage" --format json
python -m sdetkit kits expand --goal "add more helpful automation bots and GitHub Advanced Security coverage" --format json
python -m sdetkit maintenance --include-check github_automation_check --format md
python -m sdetkit maintenance --include-check github_automation_check --format json | jq '.checks.github_automation_check.details.ghas_update_tracks'
python -m sdetkit agent templates run dependency-radar-worker
python -m sdetkit agent templates run worker-alignment-radar

What this unlocks for the repo

These bots make the repo more than a set of commands; they turn it into a continuous maintenance system with:

  • better GHAS visibility,
  • campaign-oriented GHAS backlog planning,
  • age-based GHAS SLA enforcement,
  • machine-readable GHAS metrics snapshots for dashboards and audits,
  • CodeQL hotspot batching instead of one-alert-at-a-time triage,
  • coverage checks for newer GHAS operating-model features,
  • workflow-governance audits for permissions, SHA pinning, and manual recovery,
  • scheduled dependency prioritization,
  • a docs information-architecture control loop instead of ad hoc nav cleanup,
  • a recurring release-ops radar for trust-asset and workflow freshness,
  • a worker-alignment control loop that keeps template automation synchronized with the repo's strongest maintenance lanes,
  • clearer issue-driven follow-up,
  • smaller validation loops for upgrades and refactors,
  • less hidden drift between security posture and maintenance work.