CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

aws-investigation

Investigates what is happening in an AWS account from evidence already recorded there. Not for capturing new data from running code (snapshots of function arguments on a live service, Dynamic Instrumentation), questions answered from the CloudWatch Omni context graph or Application Signals service map, or building or changing infrastructure. Covers reading deployment records instead of the newest stored version (completed AppConfig deployments, the version an alias points to, a stack's current template), following a request through every service's own logs (exact per-layer fields, unique failed requests in an alarm's breach window, failure boundary per event), and answering questions from the account's data (exact counts and breakdowns, coverage of the requested range, exact filters and joins, the rule behind failed or unmatched records, verbatim values). Use to triage an alarm, error spike or failed request, trace a request across services, or count, break down, extract or join records.

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

The canonical home for this skill is aws-investigation in aws/agent-toolkit-for-aws

SKILL.md
Quality
Evals
Security

Quality

Content

92%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.

An exemplary progressive-disclosure structure: a lean overview body with a clear routing table, universally applicable rules, and an explicit closing checklist that doubles as validation. The single soft spot is that the body's own guidance, while concrete in direction, defers all copy-paste-ready specifics to the reference files.

DimensionReasoningScore

Conciseness

The ~55-line body is lean and assumes competence: no basic AWS concepts are explained, and every token carries non-obvious value ("airport-code regions such as PDX = us-west-2, IAD = us-east-1", the "claim <- artifact, record or query" provenance notation). The overlap between the rules and the closing checklist is functional validation rather than padding, so it fits anchor 5 rather than anchor 4's "minor instances that could be trimmed".

5 / 5

Actionability

The body gives mostly concrete, executable direction — a question-to-reference routing table, "list resources, log groups and data stores in the likely regions and search for the identifiers", "Use aggregate queries run in the engine" — but the fully executable specifics (query syntax, tooling, per-service field names) are deferred to the reference files rather than present in the body. For an instruction-only skill this is acceptable structure, yet it leaves minor gaps versus the copy-paste-ready level of anchor 5.

4 / 5

Workflow Clarity

A clear three-stage sequence (pick the reference → apply the five rules during investigation → close with the pre-submission checklist) with explicit validation steps and a feedback loop: "have you reread your conclusion against each piece of evidence you collected and removed any statement that contradicts it". The skill is read-only investigation, so the destructive/batch cap does not apply; the checklist-for-complex-process pattern matches anchor 5.

5 / 5

Progressive Disclosure

The body is a concise overview that routes to four real, one-level-deep reference files via a well-signaled table ("The question is about | Read"), with all linked paths verified to exist and no nested references inside the reference files themselves. Universal rules stay inline while domain detail is split out — a direct match for anchor 5.

5 / 5

Total

19

/

20

Passed

Description

92%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, third-person description that clearly states what the skill does, when to use it with concrete trigger phrases, and — unusually well — what it is explicitly not for. The only weakness is a handful of missing natural synonyms ("root cause", "incident", "debug") that users might say when they need this skill.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — "reading deployment records instead of the newest stored version", "following a request through every service's own logs", "answering questions from the account's data (exact counts and breakdowns, coverage of the requested range, exact filters and joins)" — with comprehensive coverage of its stated scope in third-person voice. It exceeds anchor 4 because there are no minor gaps: each coverage area is enumerated with concrete sub-capabilities.

5 / 5

Completeness

It explicitly answers "what" ("Investigates what is happening in an AWS account from evidence already recorded there", with three enumerated coverage areas) and "when" ("Use to triage an alarm, error spike or failed request, trace a request across services, or count, break down, extract or join records"), plus explicit out-of-scope boundaries. This matches the anchor-5 example structure of concrete capabilities followed by explicit trigger phrases.

5 / 5

Trigger Term Quality

Good natural-language coverage: "triage an alarm, error spike or failed request", "trace a request across services", "count, break down, extract or join records", plus AWS-specific terms (AppConfig, CloudWatch, alias, stack template). It falls short of anchor 5 because common investigation phrasings users would naturally say — "root cause", "incident", "debug" — are absent.

4 / 5

Distinctiveness Conflict Risk

The explicit exclusion clause — "Not for capturing new data from running code (snapshots of function arguments on a live service, Dynamic Instrumentation), questions answered from the CloudWatch Omni context graph or Application Signals service map, or building or changing infrastructure" — actively fences off adjacent capabilities, giving a clear niche with minimal wrong-skill trigger risk.

5 / 5

Total

19

/

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-core/skills/aws-investigation/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.