CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

aws-iam-passrole-chain

AWS IAM privilege escalation via `iam:PassRole` chains — Lambda/Glue/Sagemaker/EC2/ECS PassRole to a higher-priv role, AssumeRole chains across accounts, sts:GetCallerIdentity recon, account hijack via legacy root-mfa-bypass.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./packages/decepticon/decepticon/skills/standard/cloud/aws-iam-passrole-chain/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

76%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 extremely lean, command-first runbook with an unusually executable Lambda exfiltration path and good section organization. Weaknesses are procedural: validation checkpoints between phases are missing, there is no error-recovery guidance for the common failure modes, and a couple of Phase 1 commands contain ARN/version assumptions that will fail on real accounts.

Suggestions

Insert explicit verification checkpoints between phases, e.g. confirm the iam:PassRole statement's Resource matches the chosen target role ARN before create-function, and inspect the Lambda result/logs (`aws logs get-log-events`) when invoke returns an error or times out.

Add brief error-recovery guidance for the dominant failure mode: PassRole being denied on the chosen role (function creation fails with an unauthorize/PassRole error) — pick the next enumerated target rather than retrying.

Fix Phase 1 commands: resolve the policy's DefaultVersionId via `aws iam get-policy` instead of hardcoding `--version-id v1`, and use the customer-policy ARN namespace `arn:aws:iam::ACCT:policy/name` instead of `arn:aws:iam::aws:policy/`.

DimensionReasoningScore

Conciseness

The body is almost entirely executable commands, a dense permission table, and terse inline comments ("# who are you?", "# SSH into it; whoami is TARGET_ROLE") — zero concept explanation and no padding. This matches anchor 5 ("Lean and efficient; assumes Claude's competence; every token earns its place"); it is not 4 because there are no over-explained passages to trim.

5 / 5

Actionability

Mostly copy-paste-ready: the Lambda path is complete end-to-end (heredoc payload, zip, create-function, invoke, read output) and EC2/Glue/SageMaker snippets are concrete. It falls to anchor 4 ("minor gaps") rather than 5 because Phase 1's policy command assumes `--version-id v1` without resolving DefaultVersionId, uses the AWS-managed-policy ARN namespace `arn:aws:iam::aws:policy/` where a customer policy ARN is meant, and the `cut -d/ -f2` extraction breaks on role paths.

4 / 5

Workflow Clarity

The phases are clearly sequenced (Enumerate → identify PassRole targets → execute as higher-priv role → cross-account chains → privesc classes → OPSEC), but validation checkpoints are implicit at best — no step verifies that PassRole's Resource matches the target ARN before create-function, and no error recovery exists for the most common failure (invoke failing or timing out). This matches anchor 3 ("sequence present but checkpoints missing or implicit"); it is not 4 because key checkpoints are absent rather than minor, and the state-changing create operations lack verification guidance, consistent with the destructive/batch cap.

3 / 5

Progressive Disclosure

A single well-organized file with clear section headers, no nested references, and a signaled external tool pointer (Pacu link); the References section is compact. This matches anchor 4 ("Good structure; most content is appropriately placed; minor organization gaps") — it is not 5 because at ~120 lines the Phase 4 permission table and reference detail could live in a one-level-deep references/ file for a leaner main skill, and the under-50-line simple-skill exception does not apply.

4 / 5

Total

16

/

20

Passed

Description

75%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 dense, highly specific description that names concrete services, API actions, and escalation classes with strong natural trigger terms and minimal conflict risk. Its main defect is the absence of any explicit "Use when..." trigger guidance, which caps completeness at 3; it also references "account hijack via legacy root-mfa-bypass", which the skill body never covers.

Suggestions

Add an explicit trigger clause, e.g. "Use when assessing AWS IAM privilege-escalation paths in an authorized engagement and the user mentions PassRole, AssumeRole, IAM escalation, or role pivoting."

Include natural synonyms users say in this domain: "priv esc"/"privesc", "IAM enumeration", "role pivot", "credential compromise".

Remove or cover the "account hijack via legacy root-mfa-bypass" claim so the description stays aligned with what the body actually teaches.

DimensionReasoningScore

Specificity

The description lists multiple concrete, specific actions — "Lambda/Glue/Sagemaker/EC2/ECS PassRole to a higher-priv role", "AssumeRole chains across accounts", "sts:GetCallerIdentity recon", "account hijack via legacy root-mfa-bypass" — with named services and API-level verbs, not vague language. This matches the anchor "Lists multiple specific concrete actions; comprehensive coverage"; it is not score 4 because coverage of the skill's escalation paths is comprehensive rather than having minor gaps, and not below because nothing is generic.

5 / 5

Completeness

The "what" is explicit and detailed, but there is no "Use when..." clause or any equivalent explicit trigger guidance, so completeness is capped at 3 per the judging guideline. It matches anchor 3 ("Has a clear 'what' but 'when' is missing or only weakly implied") and cannot be 4 because no usage context is stated at all.

3 / 5

Trigger Term Quality

Strong domain keywords a user would naturally say: "AWS IAM privilege escalation", "PassRole", "AssumeRole", "Lambda", "Glue", "Sagemaker", "EC2", "ECS", "role chain", "GetCallerIdentity". It falls just short of anchor 5's "comprehensive coverage of natural terms including synonyms" because common variations like "priv esc"/"privesc", "IAM enumeration", or "role pivot" are absent.

4 / 5

Distinctiveness Conflict Risk

A clear niche (IAM PassRole/AssumeRole privilege-escalation chains) with distinct, unambiguous triggers; unlikely to fire for unrelated AWS skills. Matches anchor 5 ("Clear niche with distinct triggers; minimal conflict risk") and is not score 4 because the PassRole-chain framing is unique enough to avoid even minor overlap with broader AWS-enumeration skills.

5 / 5

Total

17

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository

PurpleAILAB/Decepticon

Path

packages/decepticon/decepticon/skills/standard/cloud/aws-iam-passrole-chain/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.