agentleFS
karim-bhalwanikarim-bhalwani/agentic-harness/skills/ops/SKILL.md

ops

Skill0 starsChanged 4 months agoMITView on GitHub

CI/CD pipelines, GitHub Actions workflows, Docker configurations, deployment automation, and release management patterns. Use when building CI/CD, writing GitHub Actions, creating Docker configs, automating deployments, managing releases, or setting up infrastructure-as-code. DO NOT USE FOR: writing application-specific business logic (use implementer), designing high-level system architecture or topology (use architect), performing security audits (use guardian), or orchestrating data pipelines like Airflow DAGs (use data-engineering).

17 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use status, design and 3 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • AnthropicSkill

    status

    Case status summary by audience — client-facing (plain language), internal (for the professor), or court-ready (formal caption format per local rules). Same facts, different framing and depth. Use when a student needs to update the client, brief the professor, or prepare a court status report.

    Also in 3 projects that use this

  • Microsoft AzureSkill

    design

    Design rules and conventions for the API Center Portal. Use when editing detail pages, card components, badges, layout, navigation, sidebars, or headers. Covers badge styling, metadata placement, page navigation patterns, and layout alignment.

    Also in 3 projects that use this

  • BlockSkill

    plan

    Draft a TDD, PRD, or lightweight plan in the conversation without committing it.

    Also in 3 projects that use this

  • SentrySkill

    review

    Fixture review skill.

    Also in 3 projects that use this

  • VercelSkill

    setup

    Set up Vercel CLI and project configuration. Use when the user says "set up Vercel", "configure Vercel", "link to Vercel", or "vercel init".

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 14 edited versions we read keep these.

  • 8

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Immutable artifacts: every build produces a versioned, unchangeable artifact

    line 21
  2. Distinctive:

    Secrets in vaults: never in code, config, or CI logs

    line 23
  3. Distinctive:

    Rollback before deploy: document rollback plan BEFORE deployment

    line 24
  4. Distinctive:

    Action: write a mechanical check (pre-commit hook, lint rule, or CI job)

    line 33
  5. Distinctive:

    Does NOT write application business logic (pipelines and automation only)

    line 93
  6. Distinctive:

    Does NOT manage secrets directly (provides vault integration patterns)

    line 94
  7. Distinctive:

    Does NOT perform security audits (consult guardian)

    line 95
  8. Distinctive:

    Does NOT design system architecture (consult architect)

    line 96

“Common” is measured: kept by at least 6 in 10 of the 14 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Ops Skill - CI/CD, Deployment & Automation

Version: 9.0 | Updated: 01-July-2026 | Architect: Karim Bhalwani | Tiered: core (~140 lines) + on-demand references

Unified reference for operational automation. For copy-ready templates (GitHub Actions, Docker, Terraform, release checklists), load the deep-dive reference.

Core Principles

  • Immutable artifacts: every build produces a versioned, unchangeable artifact
  • Environment parity: dev/staging/prod from same templates
  • Secrets in vaults: never in code, config, or CI logs
  • Rollback before deploy: document rollback plan BEFORE deployment

Mechanical Enforcement

When a behavioral rule fails twice in retrospectives, promote it to a mechanical check. Linters and CI are more reliable than prompt instructions.

Promote-to-Code Decision Rule

  1. Trigger: behavioral rule caused rework 2+ times
  2. Action: write a mechanical check (pre-commit hook, lint rule, or CI job)
  3. Record: log in .copilot/context/DECISIONS.md as an ADR
  4. Remove redundancy: move the original prompt instruction verbatim into .copilot/context/DECISIONS.md under the relevant ADR entry, then delete it from the prompt.

Enforcement Layers

LayerCatchesSpeedScope
Pre-commit hooksStyle, structure, obvious violationsSecondsPer-commit
CI workflowCross-file issues, Bible freshness, doc stalenessMinutesPer-PR
Guardian reviewDesign quality, logic correctness, architectural fitMinutesPer-PR

Load mechanical-enforcement.md for copy-ready pre-commit, ruff, and CI config templates.

Quick Reference: Key Patterns

GitHub Actions

  • Standard CI: lint > test > security scan (parallel where possible)
  • Reusable workflows via workflow_call for DRY deployment
  • Matrix strategy for cross-version/cross-OS testing
  • Cache pip/npm dependencies with actions/cache@v4

Docker

  • Multi-stage builds, pin base image versions, non-root user
  • .dockerignore for build context optimization
  • Health checks for orchestrated deployments

Deployment

  • Azure: OIDC via azure/login@v2, deployment slots for zero-downtime
  • Databricks: Asset Bundles for pipeline deployment
  • Blue-green or canary for stateless services
  • Database migrations: forward-only with backward compatibility

Release Management

  • Semantic versioning: MAJOR.MINOR.PATCH
  • Changelog in Keep a Changelog format
  • Release checklist: CI green, changelog updated, version bumped, security scan clean, staging tested, rollback documented, tag pushed

Infrastructure as Code

  • Terraform: modules for reuse, remote state backend, environment variables
  • Idempotent: running twice produces same result

For full templates and examples of all patterns above, load cicd-patterns-deep-dive.md.

Definition of Done

  • CI pipeline runs green (lint, test, security scan)
  • Docker image builds and runs with health check passing
  • Rollback plan documented before deployment
  • Secrets managed via vault/env vars (none in code)
  • Release checklist completed
  • Mechanical enforcement in place for P1 invariants

Constraints

  • Does NOT write application business logic (pipelines and automation only)
  • Does NOT manage secrets directly (provides vault integration patterns)
  • Does NOT perform security audits (consult guardian)
  • Does NOT design system architecture (consult architect)

Common Pitfalls

  • Over-complex pipelines: start simple, add stages only when justified
  • Missing rollback plan: every deployment needs one
  • Secret leakage: never echo secrets in logs
  • Flaky tests in CI: quarantine, don't ignore
  • Manual steps in "automated" pipelines: if it requires human intervention, it's not automated

Integration Points

  • guardian: Reviews CI/CD configurations for security and quality
  • release-manager: Consumes pipeline outputs for release decisions
  • architect: Provides infrastructure architecture and deployment topology
  • data-engineering: Deploys data pipelines via Airflow/Databricks workflows

References

Load on demand for specific sub-tasks. If a referenced file is not available, generate a minimal inline template based on the Quick Reference patterns above and note that the full template should be added to the references directory.

  • cicd-patterns-deep-dive.md - Full GitHub Actions templates (CI, reusable workflows, matrix, caching, PR automation), Docker patterns, deployment configs (Databricks, Azure), release management (semver, changelog, checklist), IaC (Terraform). Load when building or modifying pipelines.
  • mechanical-enforcement.md - Pre-commit hooks, ruff rules, GitHub Actions enforcement workflows.

Scripts

  • check_pipeline_health.py - CI/CD config validator (hard-coded secrets, unpinned images, deprecated runners, missing timeouts).
  • lint_agent_legible_errors.py - Linter enforcing agent-legible error messages with remediation hints.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.