Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Strangler Fig Orchestrator Skill · a5c-ai | DecimalAI

a5c-ai/strangler-fig-orchestrator

SkillScore73Maintained

Orchestrate strangler fig pattern implementation for gradual migration with traffic routing and cutover management

DevOpsby a5c-ai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

DevOps & SRE pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/a5c-ai-strangler-fig-orchestrator/SKILL.md

decimalai skills pull a5c-ai-strangler-fig-orchestrator --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/a5c-ai-strangler-fig-orchestrator/SKILL.md https://app.decimal.ai/s/a5c-ai-strangler-fig-orchestrator/SKILL.md

Raw file: /s/a5c-ai-strangler-fig-orchestrator/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-strangler-fig-orchestrator.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 56d ago

—+18% tokens0% turns73% of whole cases pass with skill (16/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	12%	0%
case-02	✗→✓	▲ Improved	-1%	0%
case-07	✗→✓	▲ Improved	76%	0%
case-08	✗→✓	▲ Improved	7%	0%
case-09	✗→✓	▲ Improved	45%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/31/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

a5c-ai / babysitter

library/specializations/code-migration-modernization/skills/strangler-fig-orchestrator/SKILL.md

View on GitHub ↗

License MITImported 7/31/2026Version v1

Skill contents

SKILL.md

Strangler Fig Orchestrator Skill

Orchestrates the strangler fig pattern implementation for gradual migration from legacy to modern systems with traffic routing and cutover management.

Purpose

Enable incremental migration for:

Traffic routing configuration
Feature flag management
Gradual cutover orchestration
Rollback coordination
Legacy component sunset

Capabilities

1. Traffic Routing Configuration

Configure API gateway rules
Set up routing weights
Handle header-based routing
Manage path-based routing

2. Feature Flag Management

Create feature flags
Control rollout percentage
Manage user segments
Handle A/B testing

3. Gradual Cutover Orchestration

Plan cutover phases
Execute incremental shifts
Monitor health metrics
Coordinate teams

4. Rollback Coordination

Define rollback triggers
Automate rollback
Preserve state
Document recovery

5. Progress Tracking

Track migration progress
Monitor adoption rates
Report on status
Visualize timeline

6. Legacy Component Sunset

Plan decommissioning
Verify no traffic
Archive data
Clean up resources

Tool Integrations

| Tool | Purpose | Integration Method | |------|---------|-------------------| | API Gateways | Traffic routing | Config/API | | LaunchDarkly | Feature flags | API | | Split.io | Feature management | API | | Consul | Service mesh | API | | Istio | Traffic management | Config |

Output Schema

json
{
  "orchestrationId": "string",
  "timestamp": "ISO8601",
  "migration": {
    "legacy": {
      "system": "string",
      "endpoints": []
    },
    "modern": {
      "system": "string",
      "endpoints": []
    }
  },
  "routing": {
    "strategy": "string",
    "rules": [],
    "currentWeights": {}
  },
  "progress": {
    "phase": "string",
    "percentMigrated": "number",
    "remainingEndpoints": []
  },
  "rollback": {
    "available": "boolean",
    "lastCheckpoint": "string"
  }
}

Integration with Migration Processes

monolith-to-microservices: Gradual extraction
legacy-decommissioning: Sunset planning

Related Skills

api-compatibility-analyzer: Contract verification

Related Agents

strangler-implementation-agent: Implementation
microservices-decomposer: Service extraction

Illustrative Want to see what a5c-ai/strangler-fig-orchestrator would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/1/2026, 01:49 UTC · 22 cases

Measured on 21 comparable cases — below the 22 needed to publish a lift, so this result is shown here but not on the registry card or leaderboard.

Lift

+73%

73% vs 0%

Agent turns

0%

1.0 vs 1.0

Tokens

+18%

2,191.24 vs 1,853.62

fail → pass · 16fail → fail · 6

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 We are shifting traffic from our old monolith user service to the modern microservice using an Istio service mesh. Set up the strangler pattern migration plan and provide a JSON response detailing the orchestration metadata, legacy and modern endpoint lists, current traffic split weights, migration phase with completion percentage, remaining endpoints, and rollback availability status.	fail→pass	9,987	8,625	-14%	1	1	0%	2,423	2,721	+12%	0	0	—
case-02 I need to coordinate a gradual cutover of our billing API from legacy infrastructure to a new cloud-native system using LaunchDarkly feature flags. Generate the migration orchestration output showing orchestration ID, timestamp, legacy vs modern system mappings, routing strategy and rules, current progress phase with percent migrated, and state of rollback availability with its last checkpoint.	fail→pass	12,671	6,452	-49%	1	1	0%	2,164	2,134	-1%	0	0	—
case-03 Please orchestrate the incremental migration of our inventory catalog from the legacy backend to the new GraphQL service. Provide a structured report containing the orchestration identifier, legacy and modern system endpoint details, path routing rules and current traffic weights, current phase progress with percentage completed, list of remaining endpoints, and rollback readiness.	fail→fail	12,007	22,314	+86%	1	1	0%	2,436	2,634	+8%	0	0	—
case-04 Write an SQL DDL script to add a foreign key constraint and migrate data from the legacy 'orders' table to the new 'customer_orders' table in PostgreSQL.	fail→fail	14,643	10,930	-25%	1	1	0%	2,518	2,697	+7%	0	0	—
case-05 Refactor this legacy Python Flask controller method into a standalone domain service function following clean architecture principles.	fail→fail	14,840	12,571	-15%	1	1	0%	3,015	2,899	-4%	0	0	—
case-06 Configure Let's Encrypt TLS cert-manager annotations on an NGINX Ingress manifest for auto-renewing SSL certificates.	fail→fail	13,188	11,066	-16%	1	1	0%	2,204	2,502	+14%	0	0	—
case-07 We are migrating user authentication traffic. Beta testers with header 'X-Beta-User: true' should route to modern auth-v2 service while default traffic stays on legacy auth-v1. Output the migration JSON configuration.	fail→pass	7,801	7,773	-0%	1	1	0%	1,118	1,973	+76%	0	0	—
case-08 Shift 20% of payment gateway traffic to the modern modern-pay microservice while keeping 80% on legacy-pay. Provide the migration state JSON.	fail→pass	8,410	6,705	-20%	1	1	0%	1,778	1,905	+7%	0	0	—
case-09 Define a rollback checkpoint for the search service migration where HTTP 5xx error rate exceeding 2% triggers automated reversion to legacy-search. Return the orchestration JSON.	fail→pass	18,141	7,968	-56%	1	1	0%	1,606	2,328	+45%	0	0	—
case-10 Set up a 10% progressive canary rollout for modern recommendation service using feature flags before full cutover. Provide the orchestration JSON output.	fail→pass	24,545	9,416	-62%	1	1	0%	1,856	2,252	+21%	0	0	—
case-11 We have reached 100% traffic on modern-notification-service. Prepare the final phase orchestration payload for decommissioning the legacy notification monolith endpoints.	fail→pass	13,797	8,501	-38%	1	1	0%	2,691	2,226	-17%	0	0	—
case-12 Route requests matching path `/api/v2/catalog/` to modern-catalog while `/api/v1/catalog/` remains on legacy-monolith. Generate the JSON orchestration state.	fail→fail	4,676	6,399	+37%	1	1	0%	960	2,077	+116%	0	0	—
case-13 Advance the order service strangler migration to phase 2 with a 50/50 split between legacy-orders and modern-orders. Output the JSON schema response.	fail→pass	8,675	6,236	-28%	1	1	0%	1,896	1,937	+2%	0	0	—
case-14 Route internal employee traffic with header 'X-Internal-Employee: 1' to modern-hr-service during initial phase 1 cutover. Generate the migration payload.	fail→pass	19,899	8,732	-56%	1	1	0%	1,067	2,527	+137%	0	0	—
case-15 Configure an A/B test routing setup where 15% of checkout traffic lands on modern-checkout-v2 and 85% stays on legacy-checkout. Provide the strangler JSON response.	fail→pass	7,346	8,077	+10%	1	1	0%	1,458	2,415	+66%	0	0	—
case-16 Create an initial baseline strangler orchestration setup for legacy monolith shipping service prior to any live traffic shift. Output JSON.	fail→pass	14,560	5,756	-60%	1	1	0%	3,067	1,871	-39%	0	0	—
case-17 Set up an active rollback checkpoint 'cp-2023-10-15-db-sync' for the customer profile service migration. Return JSON orchestration output.	fail→pass	8,869	7,707	-13%	1	1	0%	1,326	1,950	+47%	0	0	—
case-18 Update strangler migration for accounting API to phase 3 with 75% traffic on modern-accounting and 25% on legacy-accounting. Output JSON payload.	fail→pass	4,296	4,369	+2%	1	1	0%	841	1,578	+88%	0	0	—
case-19 Configure path routing so `/media/stream/*` routes to modern-streamer and remaining paths stay on legacy-media-server. Generate orchestration JSON.	fail→fail	7,292	9,003	+23%	1	1	0%	1,421	1,971	+39%	0	0	—
case-20 Migrate `/inventory/stock` to modern service while `/inventory/warehouse` remains on legacy monolith. Generate strangler migration status JSON.	fail→pass	6,490	7,209	+11%	1	1	0%	1,363	2,195	+61%	0	0	—
case-21 Prepare orchestration payload for final pre-sunset phase where modern-auth handles 90% of traffic and legacy-auth handles 10%. Output JSON.	fail→pass	7,991	6,280	-21%	1	1	0%	1,705	1,993	+17%	0	0	—
case-22 Direct requests with header 'User-Agent: MobileApp/2.0' to modern-mobile-api while default traffic remains on legacy-api. Output migration state JSON.	fail→pass	5,536	5,965	+8%	1	1	0%	1,080	1,758	+63%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +73 percentage points is the difference between those two pass rates over the 21 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 0%
Pass rate with the skill: 72.7%
Cases attempted: 22
Cases counted toward lift: 21
Cases that regressed: 0
Skill version measured: v1
Run completed: 8/1/2026, 01:49 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-01

Prompt

We are shifting traffic from our old monolith user service to the modern microservice using an Istio service mesh. Set up the strangler pattern migration plan and provide a JSON response detailing the orchestration metadata, legacy and modern endpoint lists, current traffic split weights, migration phase with completion percentage, remaining endpoints, and rollback availability status.

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run73

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 55 days ago

Other measured skills in the registry, with their headline benchmark lift.