hunt-cloud-misconfig

Hunt cloud / infrastructure misconfigurations. AWS: public S3 buckets (s3:GetObject anonymous), permissive bucket policies (PutObjectAcl public-write), exposed CloudFront origin, public Lambda function URL, public RDS snapshot, IAM credentials in JS bundles, AWS metadata accessible via SSRF. GCP: public GCS buckets, exposed Cloud Run services, leaked service account JSON. Azure: public blob containers, exposed Function App. (Kubernetes/Docker exposure is owned by hunt-k8s; CI/CD pipeline attacks by hunt-cicd; post-credential IAM escalation by cloud-iam-deep.) Detection: targeted dorking, certificate transparency, JS bundle secret extraction, port scan for known service ports. Validate: actual data read / write / RCE. Use when hunting cloud-native storage and compute misconfig (S3/GCS/Blob, IMDS-via-SSRF, serverless, public managed services).

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Quality

Content

81%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.

Highly actionable content with executable commands, regex sets, and a strong validation checklist including destructive-operation safeguards. Its main weakness is conciseness: the RUM deep-dive, severity table, and dated writeup list add prose that could move to a referenced file.

Suggestions

Move the CloudWatch RUM severity rubric and 'authoritative writeups' list into a one-level-deep reference file (e.g. references/rum-deepdive.md) and link it from the body to cut inline tokens.

Trim explanatory prose around the RUM attack chains (e.g. 'AWS-owned domain on every enterprise allowlist... DLP doesn't parse...') to command-and-result brevity.

Relocate dated/version-pinned material (LocalStack 3.0 vs 4.x, awscli ≥ 2.30 workaround) into a clearly labeled 'version notes' or 'deprecated/old patterns' section so time-sensitive content does not inflate the main flow.

Dimension	Reasoning	Score
Conciseness	Code and commands are lean, but the CloudWatch RUM section carries substantial prose explanation, a dated severity rubric table, and a long 'authoritative writeups' list that pads tokens; it is mostly efficient but includes unnecessary explanation that could be tightened — not level 4's 'minor instances'.	3 / 5
Actionability	Copy-paste ready curl/aws/grep commands, a complete regex set, and full attack-chain bash blocks cover the common cases concretely — fully executable, matching level 5.	5 / 5
Workflow Clarity	A 7-step validation checklist with explicit checkpoints and a read-only-PoC gate ("Do not modify/delete data even if permitted") plus a LocalStack verification flow with a documented env-var workaround provides clear sequencing with feedback loops.	5 / 5
Progressive Disclosure	Sections are well-organized and self-contained (no bundle files exist, so no nested references), but the deep RUM material — severity rubric and writeup list — is inlined where a one-level reference would help; good structure with minor organization gaps, not the cleanly split level-5.	4 / 5
	Total	17 / 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, well-bounded description that names concrete actions per cloud provider, explicitly states when to use it, and carves out adjacent skills to avoid conflicts. Its only weakness is a jargon-heavy register that slightly limits natural trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete per-cloud actions ("public S3 buckets (s3:GetObject anonymous)", "permissive bucket policies (PutObjectAcl public-write)", "public Lambda function URL", "IAM credentials in JS bundles", "AWS metadata accessible via SSRF") plus detection and validation primitives — comprehensive coverage, not level 4's 'minor gaps'.	5 / 5
Completeness	Explicitly answers both what ("Hunt cloud / infrastructure misconfigurations") and when ("Use when hunting cloud-native storage and compute misconfig (S3/GCS/Blob, IMDS-via-SSRF, serverless, public managed services)") with concrete trigger phrases — a clear level-5 match, not the weaker 'when' of level 4.	5 / 5
Trigger Term Quality	Strong natural terms ("cloud misconfig", "S3/GCS/Blob", "serverless", "public managed services") but the description leans heavily on technical jargon and is missing a few common user-facing synonyms; below the comprehensive synonym-plus-extension coverage of level 5.	4 / 5
Distinctiveness Conflict Risk	Explicit boundary carve-outs ("Kubernetes/Docker exposure is owned by hunt-k8s; CI/CD pipeline attacks by hunt-cicd; post-credential IAM escalation by cloud-iam-deep") give it a clear niche with minimal conflict risk.	5 / 5
	Total	19 / 20 Passed

Validation

87%

Warnings & errors only

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

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 missing	Warning

	Total	14 / 16 Passed

Repository: elementalsouls/Claude-BugHunter
Path: skills/hunt-cloud-misconfig/SKILL.md
Commit: d4b54e1

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

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.