Ready to supercharge your AI agents?

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

Detecting AWS IAM Privilege Escalation · mukul975 | DecimalAI

Detecting AWS Iam Privilege Escalation

SkillScore5Maintained

detecting-aws-iam-privilege-escalation

Detect AWS IAM privilege escalation paths using boto3 and Cloudsplaining policy analysis to identify overly permissive policies, dangerous permission combinations, and least-privilege violations

Securityby mukul975

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

Automatic

Security & AppSec pack

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

pip install decimalai

2 · Get the skill — writes .claude/skills/detecting-aws-iam-privilege-escalation/SKILL.md

decimalai skills pull detecting-aws-iam-privilege-escalation --out .claude/skills/

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

curl -fsSL --create-dirs -o .claude/skills/detecting-aws-iam-privilege-escalation/SKILL.md https://app.decimal.ai/s/detecting-aws-iam-privilege-escalation/SKILL.md

Raw file: /s/detecting-aws-iam-privilege-escalation/SKILL.md ↗ · machine-readable summary at /s/detecting-aws-iam-privilege-escalation.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 55d ago

Yes — +5% lift+5% tokens0% turns5% of whole cases pass with skill (1/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-13	✗→✓	▲ Improved	—	—
case-10	✗→✗	＝ Same ✗	—	—
case-01	✗→✗	＝ Same ✗	—	—
case-19	✗→✗	＝ Same ✗	—	—
case-03	✗→✗	＝ Same ✗	—	—

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/30/2026

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

mukul975 / Anthropic-Cybersecurity-Skills

skills/detecting-aws-iam-privilege-escalation/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 8/1/2026Version v1Bundled 2 files · scripts

Skill contents

scripts/

references/

SKILL.md

Detecting AWS IAM Privilege Escalation

Overview

This skill uses boto3 and Cloudsplaining-style analysis to identify IAM privilege escalation paths in AWS accounts. It downloads the account authorization details, analyzes each policy for dangerous permission combinations (iam:PassRole + lambda:CreateFunction, iam:CreatePolicyVersion, sts:AssumeRole), and flags policies that violate least-privilege principles.

When to Use

When investigating security incidents that require detecting aws iam privilege escalation
When building detection rules or threat hunting queries for this domain
When SOC analysts need structured procedures for this analysis type
When validating security monitoring coverage for related attack techniques

Prerequisites

Python 3.8+ with boto3 library
AWS credentials with IAM read-only access (iam:GetAccountAuthorizationDetails)
Optional: cloudsplaining Python package for HTML report generation

Steps

Download IAM Authorization Details — Call iam:GetAccountAuthorizationDetails to retrieve all users, groups, roles, and policies
Analyze Policies for Privilege Escalation — Check each policy for known escalation permission combinations
Identify Wildcard Resource Policies — Flag policies using Resource: "" with dangerous actions
Map Principal-to-Policy Relationships — Build a graph of which principals can access which escalation paths
Score and Prioritize Findings — Rank findings by severity based on escalation vector type
Generate Report — Produce structured JSON report with remediation guidance

Expected Output

JSON report of privilege escalation findings with severity scores
List of dangerous permission combinations per principal
Wildcard resource policy audit results
Remediation recommendations for each finding

Illustrative Want to see what detecting-aws-iam-privilege-escalation 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

7/28/2026, 01:58 UTC · 22 cases

Lift

+5%

5% vs 0%

Agent turns

1.0 vs 1.0

Tokens

+5%

2,289.73 vs 2,183.77

fail → pass · 1fail → fail · 21

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-10 When generating an IAM privilege escalation assessment, why is analyzing policy documents in isolation insufficient compared to mapping principal-to-policy relationships?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-01 We want to set up real-time streaming alerts using CloudTrail and EventBridge whenever an unauthorized user attempts an IAM privilege escalation call in production. Outline the CloudTrail event pattern and Lambda trigger architecture for this real-time detection pipeline.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-19 If a low-privileged IAM principal possesses `iam:CreateAccessKey` targeted at another IAM user, how can this permission be leveraged for privilege escalation?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-03 Write a Python script using boto3 that automatically deletes overprivileged IAM policies and revokes permissions in our production AWS account immediately upon identifying an escalation path.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-08 In an automated IAM evaluation script, how should policies pairing dangerous administrative actions with `Resource: "*"` be classified compared to policies restricted to explicit ARNs?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-12 What optional Python package can be integrated alongside boto3 to render interactive HTML reports detailing IAM policy privilege escalation vectors?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-07 When auditing IAM policies for escalation risks, we find `sts:AssumeRole` granted on `Resource: "*"`. Why is granting `sts:AssumeRole` with a wildcard resource flagged as a severe escalation risk in multi-role AWS accounts?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-17 Describe how combining `iam:PassRole` with `ec2:RunInstances` creates an administrative privilege escalation vector in an AWS account.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-13 We are preparing a Docker container environment to execute our AWS IAM policy analysis scripts using boto3. What minimum Python runtime version should be specified in the environment setup?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-09 What minimal AWS IAM permission must be assigned to an audit role to allow downloading all account authorization details required for Cloudsplaining-style policy analysis?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-06 An auditor notices a developer policy granting `iam:CreatePolicyVersion` on `Resource: "*"`. The auditor asserts this is low risk because the developer cannot create new IAM policies or attach policies to users. Is this permission sufficient to achieve administrator privilege escalation?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-11 We are building an automated CI/CD pipeline step that runs an IAM privilege escalation scan. What structured output format should the tool produce to enable downstream machine ingestion and remediation tracking?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-18 An IAM user has permission to execute `iam:AddUserToGroup` on `Resource: "*"`. What privilege escalation path does this grant to the user?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-20 A developer role possesses `iam:UpdateAssumeRolePolicy` targeting an internal deployment role. Explain how this permission allows the developer to elevate their access.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-05 During an IAM policy audit, a policy is flagged for allowing `lambda:CreateFunction` and `lambda:InvokeFunction` on all resources. An analyst argues this cannot be used for privilege escalation without a specific secondary permission. Which missing IAM permission combined with Lambda creation forms the primary privilege escalation vector?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-04 We need to fetch the complete state of all users, groups, roles, and managed policies in an AWS account in a single API call to perform privilege escalation analysis. A junior engineer suggests calling `iam.list_policies()` followed by `iam.list_roles()` and iterating with `get_policy`. What single IAM API operation should be used to retrieve all account authorization details at once?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-21 When synthesizing findings from an IAM authorization scan into a final assessment, how should individual policy findings be prioritized for SOC analyst review?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-16 An attacker discovers they have `iam:SetDefaultPolicyVersion` permission on a customer-managed policy attached to their role. How is this permission exploited to gain elevated privileges?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-14 An IAM policy allows `iam:AttachRolePolicy` on `Resource: "*"`. How does an attacker holding this permission achieve full administrator access without modifying existing policy documents?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-02 We are conducting a security audit of our Google Cloud Platform environment and need to detect service accounts with privilege escalation paths using GCP Cloud Asset Inventory. How should we configure our IAM policy scanner for GCP service accounts?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-15 A security review flags `iam:PutRolePolicy` on a user's own role as a critical finding. Explain the precise mechanism an attacker uses to escalate privileges with this action.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-22 What essential information must be provided alongside each privilege escalation finding in the final audit findings report to assist cloud security engineers?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—

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. The headline lift of +5 percentage points is the difference between those two pass rates over the 22 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: 4.6%
Cases attempted: 22
Cases counted toward lift: 22
Cases that regressed: 0
Skill version measured: v1
Run completed: 7/28/2026, 01:58 UTC

The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run5

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.