Skip to content
cccskills
FreePublished on GitHub

deploy

Post-merge deployment workflow covering feature flags, canary/blue-green rollout, smoke-test verification, and rollback triggers. Use after a PR is merged and you need to ship it safely to production. Trigger on "deploy", "roll out", "canary", "feature flag", "rollback".

The English summary is being prepared. Showing the original description.

See how to install

Files included (10)

  • SKILL.md5.6 KB
  • templates/canary/patterns.md2.3 KB
  • templates/canary/summary.md1.3 KB
  • templates/flags/patterns.md2.7 KB
  • templates/flags/summary.md1.2 KB
  • templates/overview.md1.3 KB
  • templates/rollback/patterns.md3.1 KB
  • templates/rollback/summary.md1.8 KB
  • templates/verify/patterns.md2.6 KB
  • templates/verify/summary.md1.5 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

Deploy — Post-Merge Deployment Workflow

Structured guidance for everything between /pr-merge (code merged) and /ops (long-term observability). Covers feature flags, gradual rollout, post-deploy verification, and rollback conditions. Auto-detects context from project signals or accepts explicit subcommand.

Usage

/deploy              # Auto-detect from project signals
/deploy flags        # Explicit: feature flag setup (LaunchDarkly, Flipt, env-based)
/deploy canary       # Explicit: canary / blue-green gradual rollout
/deploy verify       # Explicit: post-deploy smoke tests and SLO gate
/deploy rollback     # Explicit: rollback trigger conditions and strategies

Workflow

Step 1: Determine Context

Parse $ARGUMENTS for explicit mode. If no arguments, run context detector:

DETECTED=$(bash "$HOME/.claude/skills/deploy/lib/context-detector.sh")

Step 2: Show Relevant Guidance

Based on detected or explicit mode, use progressive disclosure:

Level 1 — Summary (default, ~15 lines): Read and present the summary template. This is the entry point.

Level 2 — Patterns (on request, ~60 lines): When the user asks for more detail, "show me patterns", code examples, or "Don't/Do/Best" style guidance, read the patterns template.

Level 3 — Full SOP (future): Not yet available. Dive into patterns and answer specific questions instead.

Step 3: Handle Multiple Contexts

If context-detector returns multiple modes (e.g., "verify,rollback"):

  • Show summary for each detected mode
  • Ask: "Which area do you want to dive deeper into?"

Step 4: No Context Detected

If detection returns "none" and no explicit mode was given:

  • Read and show templates/overview.md
  • Let the user choose which mode to start with

Template Locations

Summary Templates (~15 lines each — L1 default)

ModeFile
flags~/.claude/skills/deploy/templates/flags/summary.md
canary~/.claude/skills/deploy/templates/canary/summary.md
verify~/.claude/skills/deploy/templates/verify/summary.md
rollback~/.claude/skills/deploy/templates/rollback/summary.md

Patterns Templates (~60 lines each — L2 on request)

ModeFile
flags~/.claude/skills/deploy/templates/flags/patterns.md
canary~/.claude/skills/deploy/templates/canary/patterns.md
verify~/.claude/skills/deploy/templates/verify/patterns.md
rollback~/.claude/skills/deploy/templates/rollback/patterns.md

Reference Templates

ModeFile
overview~/.claude/skills/deploy/templates/overview.md

Progressive Disclosure Rules

  1. Always start at L1 — never dump 300+ lines unprompted
  2. Read templates on demand — use the Read tool to load templates, don't memorize content
  3. User drives depth — "show me patterns", "give me code", "how do I set up X" triggers L2
  4. Cross-reference, don't duplicate — for secrets in flags, cite /security operations; for post-deploy metrics, cite /ops metrics

Relationship to Other Skills

  • /pr-merge: Merge is the trigger. /deploy is the next step after the PR lands on main.
  • /ops: Covers long-term observability (logging, metrics, dashboards, alerting). /deploy verify is the immediate post-deploy gate; /ops sets up the sustained monitoring layer.
  • /security operations: Covers secrets management and deployment security. Use before configuring flag SDK credentials.
  • /implementation: After implementing a flagged feature, use /deploy flags to wire the flag evaluation into the request path.

Examples

Auto-detected flags context (launchdarkly in deps):

→ context-detector returns "flags" → Read and show templates/flags/summary.md → User: "show me the SDK patterns" → Read and show templates/flags/patterns.md

Explicit canary request:

User: /deploy canary → Read and show templates/canary/summary.md

No context, overview:

User: /deploy → context-detector returns "none" → Read and show templates/overview.md

Post-merge verification flow:

PR merged → /deploy verify → smoke tests pass → promote to 100% traffic → if any check fails → /deploy rollback

Gotchas

  • Never roll back a DB migration without a tested down-migration — data loss risk. Stop traffic and escalate instead.
  • Automatic rollback triggers on 5xx > 1% absolute, p99 > baseline × 1.5, or health endpoint non-200 — these fire without manual confirmation; know them before deploying.
  • Flag-gated and canary rollbacks are instant (flip flag / drain traffic); only a full-deploy rollback needs git revert + redeploy (5–15 min) — pick the fastest path that applies.
  • Every automatic rollback in production requires a post-mortem, and the rollback event must be logged (version numbers, trigger, timestamp).

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Validate WCAG 2.1 Level AA compliance and accessibility best practices. Use when performing accessibility audits and WCAG certification.

The English summary is being prepared. Showing the original description.

matteocervelli/llms25Updated May 22, 2026

Analyze feature requirements, dependencies, and security considerations. Use when starting feature implementation from GitHub issues to understand scope, technical feasibility, and risks.

The English summary is being prepared. Showing the original description.

matteocervelli/llms25Updated May 22, 2026

Run SQL queries against psql, BigQuery, or MySQL from the terminal, including natural-language-to-SQL and schema exploration. Use when analyzing data, inspecting DB state, or debugging tables. Trigger on "query the database", "SQL", "show me data from", "explore table".

The English summary is being prepared. Showing the original description.

matteocervelli/llms25Updated May 22, 2026

Design REST APIs or function contracts with clear request/response specifications, error handling patterns, authentication strategies, and comprehensive documentation.

The English summary is being prepared. Showing the original description.

matteocervelli/llms25Updated May 22, 2026

Generate comprehensive API endpoint tests for REST and GraphQL APIs. Creates tests for all HTTP methods, status codes, authentication, and validation.

The English summary is being prepared. Showing the original description.

matteocervelli/llms25Updated May 22, 2026

Design component architecture and module structure using established architectural patterns for clean, maintainable, and scalable systems.

The English summary is being prepared. Showing the original description.

matteocervelli/llms25Updated May 22, 2026

See all skills by matteocervelli

Report a problem with this skill