ad-coercer

Authentication coercion against Windows / AD — PetitPotam (MS-EFSR), PrinterBug (MS-RPRN), DFSCoerce (MS-DFSNM), ShadowCoerce (MS-FSRVP), Coercer.py meta-tool. Force a Windows machine to NTLM-authenticate to attacker, then relay or crack offline.

Quality

81%

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

80%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Actionable and concise reference with strong executable commands and a clear high-level workflow, held back by missing explicit validation checkpoints for destructive AD operations and a monolithic single-file structure with no progressive disclosure.

Suggestions

Add explicit validation gates between steps — e.g., after step 2 confirm ntlmrelayx logged 'Authenticating … SUCCEED' before escalating, and after ADCS relay confirm 'GOT CERTIFICATE' before running certipy — to lift workflow_clarity from 2 to 3.

Move detailed supporting material (the full vector/mitigation table, OPSEC notes, and the DCSync escalation chain) into a references file (e.g. references/coercion-vectors.md) linked from a concise overview, improving progressive_disclosure toward 3.

Dimension	Reasoning	Score
Conciseness	Lean and dense with no basic-concept padding (no explanation of what NTLM/AD is); the table and one-line summary ('All four boil down to: an unauthenticated/low-priv RPC call…') earn their tokens, matching the 'lean and efficient' anchor; KB numbers are mitigation references, not stale API time-sensitivity that the guideline penalizes.	3 / 3
Actionability	Provides fully executable, copy-paste-ready commands with real flags (impacket-ntlmrelayx, Coercer.py coerce, PetitPotam.py, certipy auth, impacket-secretsdump), matching the 'fully executable code/commands' anchor; not at 2 since there is no pseudocode or missing key details.	3 / 3
Workflow Clarity	Steps are clearly sequenced (1. listener → 2. coerce → 3. catch/escalate) but there are no explicit validation/verification checkpoints for a destructive AD-compromise operation (escalating to Domain Admins / DCSync), so per the feedback-loops guideline workflow_clarity is capped at 2; expected output is shown but not framed as a verify-before-proceeding gate.	2 / 3
Progressive Disclosure	Well-organized into clear navigable sections with a clearly signaled external References list, but it is a monolithic single-file document (~95 lines) with no file-based disclosure and substantial inline sections (OPSEC, escalation chain) that could be split; matches the 'some structure, content that should be separate is inline' anchor rather than the 'appropriately split' anchor at 3.	2 / 3
	Total	10 / 12 Passed

Description

82%

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-triggered description for a narrow AD-coercion niche, weakened only by the absence of an explicit 'Use when…' clause. Voice is appropriately third-person/imperative with no first/second-person penalty.

Suggestions

Add an explicit trigger clause such as 'Use when coercing Windows/AD authentication via PetitPotam-class RPC calls or when the user mentions NTLM relay, ADCS relay, or machine-account coercion.' to lift completeness from 2 to 3.

Dimension	Reasoning	Score
Specificity	Names concrete actions ('Force a Windows machine to NTLM-authenticate to attacker, then relay or crack offline') plus five named techniques (PetitPotam, PrinterBug, DFSCoerce, ShadowCoerce, Coercer.py) — matches the 'lists multiple specific concrete actions' anchor; not below 2 since it far exceeds 'names domain and some actions'.	3 / 3
Completeness	Clearly answers 'what' (coercion + relay/crack) but lacks any explicit 'Use when…' trigger clause, so per the judging guideline completeness is capped at 2; it cannot reach 3 without explicit 'when' guidance.	2 / 3
Trigger Term Quality	Contains the exact natural terms a user would say for this domain ('PetitPotam', 'PrinterBug', 'DFSCoerce', 'ShadowCoerce', 'Coercer', 'NTLM', 'relay') and rich when_to_use metadata (msefsr, msrprn, msdfsnm, msfsrvp); matches the 'good coverage of natural terms' anchor rather than the partial-coverage anchor at 2.	3 / 3
Distinctiveness Conflict Risk	Targets a clear niche (Windows/AD authentication coercion) with distinct technique names unlikely to trigger for unrelated skills; matches the 'clear niche with distinct triggers' anchor rather than the 'somewhat specific but could overlap' anchor at 2.	3 / 3
	Total	11 / 12 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/ad/coercer/SKILL.md
Commit: e34afba

Reviewed: about 16 hours 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.