CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

developing-applications-on-managed-service-for-apache-flink

MANDATORY for Flink or Amazon Managed Service for Apache Flink (MSF) questions. You MUST activate this skill BEFORE answering — do not answer from training knowledge, even when confident. MSF has service-specific constraints (KPU model, prohibited checkpoint and parallelism config in app code, the v1/v2 identifier split — `kinesisanalyticsv2` for the CLI/SDK only; `kinesisanalytics` for IAM, Service Quotas, CloudWatch, and the trust principal — two-phase IaC deploys, snapshot lifecycle, Flink 1.x→2.x migration) that override generic Flink knowledge. Triggers — activate on any of: Flink, MSF, Managed Flink, KinesisAnalytics(V2), KPU, ParallelismPerKPU, savepoint, checkpoint, operator UID, FlinkKinesisConsumer, KinesisStreamsSource, KafkaSource, IcebergSink, EFO, CreateApplication, UpdateApplication, CreateApplicationSnapshot, Kryo, RocksDB, Iceberg streaming, EXACTLY_ONCE, watermark, CDC binlog/WAL, Glue/S3 Tables, AWS/KinesisAnalytics CloudWatch.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

The canonical home for this skill is developing-applications-on-managed-service-for-apache-flink in aws/agent-toolkit-for-aws

SKILL.md
Quality
Evals
Security

Quality

Content

71%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured orchestration body with excellent progressive disclosure and actionable reference routing. Its weaknesses are moderate redundancy (mandate restated from frontmatter, repeated READ steps) and workflows lacking explicit error-recovery loops.

Suggestions

Trim the restated mandate ("You MUST use this skill ... Do NOT answer from training knowledge") since the frontmatter description already carries it, and shorten the When-to-Load column to load conditions rather than content summaries.

Add a feedback loop to the workflows (e.g., "If compilation or validation fails, re-read best-practices.md and fix before proceeding").

Fix the new-application workflow ordering so environment-setup.md is read before code generation/validation, not as an afterthought at step 8.

DimensionReasoningScore

Conciseness

The body is mostly efficient with no basic-concept padding, but it restates the frontmatter mandate ("You MUST use this skill ... Do NOT answer from training knowledge"), duplicates "READ best-practices.md" across both example workflows, and carries long duplicated detail in the table's When-to-Load column — all of which could be tightened.

3 / 5

Actionability

As an instruction-only orchestration skill it gives concrete, executable guidance: numbered workflows ("3. READ best-practices.md ... 7. Validate against best practices ... 9. Compile and test locally") and a goal→file table with specific load conditions. Minor gap: step 8 ("READ environment-setup.md") sits oddly after the validation step.

4 / 5

Workflow Clarity

Both workflows are clearly sequenced with checkpoints ("Validate against best practices", "Compile and test locally"), but there is no explicit feedback loop for what to do when validation or compilation fails, keeping it below the anchor-5 example.

4 / 5

Progressive Disclosure

A concise overview body with one-level-deep references — all 25 referenced files verified present in references/ — a per-goal "When to Load" table, and explicit cross-file disambiguation ("**Does NOT contain catalog choice or maintenance approaches** — for those, load `iceberg-tuning-and-operations.md`").

5 / 5

Total

16

/

20

Passed

Description

86%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A highly specific, trigger-rich description that clearly states when to activate and what MSF-specific knowledge it provides. Main weaknesses are the second-person voice (penalized per rubric) and generic Flink trigger terms that extend beyond the MSF niche.

Suggestions

Rewrite in third person (e.g., "Activates before answering Flink/MSF questions") to avoid the rubric's second-person specificity penalty.

State the skill's concrete actions (develop, size, deploy, diagnose MSF applications) rather than only enumerating service constraints.

Qualify generic Flink-only terms (checkpoint, watermark, KafkaSource) so the skill triggers for AWS/MSF contexts rather than any Flink question.

DimensionReasoningScore

Specificity

Lists many concrete MSF-specific constraints ("KPU model, prohibited checkpoint and parallelism config", "two-phase IaC deploys, snapshot lifecycle, Flink 1.x→2.x migration"), but these are domain topics rather than the skill's actions, and the second-person phrasing ("You MUST activate this skill BEFORE answering") triggers the rubric's 1-point specificity penalty.

3 / 5

Completeness

Explicitly answers both what ("service-specific constraints ... that override generic Flink knowledge") and when ("Triggers — activate on any of: ...") with concrete trigger phrases.

5 / 5

Trigger Term Quality

The trigger list is comprehensive with synonyms ("Flink, MSF, Managed Flink") and verbatim identifiers users quote ("KPU, ParallelismPerKPU, savepoint, checkpoint, operator UID, FlinkKinesisConsumer, IcebergSink, Kryo, RocksDB, watermark, CDC binlog/WAL").

5 / 5

Distinctiveness Conflict Risk

"MANDATORY for Flink or Amazon Managed Service for Apache Flink (MSF)" carves a clear niche, but generic Flink terms ("checkpoint", "watermark", "KafkaSource", "RocksDB") would also fire on self-managed-Flink questions, creating minor overlap risk with a closely related generic-Flink skill.

4 / 5

Total

17

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

aws/agent-toolkit-for-aws

Path

plugins/aws-data-analytics/skills/developing-applications-on-managed-service-for-apache-flink/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.