prototype-pollution

Hunt JavaScript prototype pollution (CWE-1321) — the 2023-2026 meta-vulnerability that chains into RCE, auth bypass, and SSRF on most Node.js stacks.

Quality

68%

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

Fix and improve this skill with Tessl

tessl review fix ./packages/decepticon/decepticon/skills/standard/analyst/prototype-pollution/SKILL.md

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.

The body is actionable and efficient with executable commands, a copy-paste PoC, and a real validation/feedback loop — strong on actionability and workflow clarity. Its main gap is progressive disclosure: everything lives in one inlined file with no reference split, and a few gadgets lack their own executable PoCs.

Suggestions

Split the sink library list and CVSS table into a reference file (e.g. references/SINKS.md) and link to it from SKILL.md to improve progressive disclosure.

Add a one-line executable PoC or payload snippet for the non-spawn gadgets (Handlebars/EJS, mongoose, lodash.template) so every gadget is reproducible.

Bind each audit-workflow step to the specific command that performs it (e.g. step 1 → the jq sweep) to tighten workflow clarity.

Dimension	Reasoning	Score
Conciseness	Mostly lean — library lists, executable commands, and a copy-paste PoC with little padding; the intro 'universal gadget' framing and a few gadget prose blurbs could be trimmed, but it generally assumes Claude's competence rather than re-explaining basics.	4 / 5
Actionability	Concrete, executable guidance throughout (jq/npm sweep commands, a copy-paste curl PoC, specific library + version names); minor gaps in that only the spawn gadget gets a full PoC while Handlebars/mongoose/etc. are described without executable examples.	4 / 5
Workflow Clarity	A clearly sequenced 4-step audit workflow plus a dedicated validation section with success signals and a negative control (the 'proto' no-underscores control), providing real feedback loops; steps are slightly high-level and not every step is bound to a specific command.	4 / 5
Progressive Disclosure	Well organized into 8 labeled sections, but it is a single ~100-line monolithic file with no external references; the CVSS table, full gadget catalog, and sink library list are candidates for separate reference files that would aid navigation.	3 / 5
	Total	15 / 20 Passed

Description

65%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.

The description is distinctive and keyword-rich for its security niche, with a clear 'what'. Its main weakness is the absence of an explicit 'Use when...' trigger clause, which caps completeness, and a single-verb action vocabulary that limits specificity.

Suggestions

Add an explicit 'Use when ...' clause naming natural trigger phrases (e.g. 'Use when auditing Node.js apps for prototype pollution, or when a chain needs an RCE/auth-bypass/SSRF first hop').

Broaden the action verb beyond 'Hunt' to name concrete performed actions (e.g. 'Hunt, confirm, and chain JavaScript prototype pollution ...').

Add common synonyms ('prototype poisoning', 'pollution gadget') to improve trigger term coverage.

Dimension	Reasoning	Score
Specificity	Names the domain ('JavaScript prototype pollution (CWE-1321)') and one concrete action ('Hunt') plus the chain outcomes (RCE, auth bypass, SSRF), but the skill's own action vocabulary is essentially a single verb with the rest being outcomes rather than performed actions.	3 / 5
Completeness	The 'what' is clear (hunt prototype pollution that chains into RCE/auth bypass/SSRF), but there is no explicit 'Use when...' clause or equivalent trigger guidance, capping completeness at 3 per the rubric guidelines.	3 / 5
Trigger Term Quality	Strong, natural security keywords a pentester would actually say — 'prototype pollution', 'RCE', 'auth bypass', 'SSRF', 'Node.js' — with good coverage; only minor natural synonyms (e.g. 'prototype poisoning', 'pollution chain') are missing.	4 / 5
Distinctiveness Conflict Risk	A clearly bounded niche (JavaScript prototype pollution, CWE-1321) with distinct triggers and minimal overlap risk against other skills.	5 / 5
	Total	15 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: PurpleAILAB/Decepticon
Path: packages/decepticon/decepticon/skills/standard/analyst/prototype-pollution/SKILL.md
Commit: 0cf691e

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.