CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

mustang-panda-bronze-president

Adversary-emulation profile for Mustang Panda (G0129 / Bronze President / Stately Taurus / RedDelta / TA416 / TEMP.Hex), a China-based state-sponsored cyber-espionage actor operating since at least 2012.

Quality

66%

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/shared/adversary-emulation/mustang-panda/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

57%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 profile is a well-structured, information-dense adversary-emulation reference with strongly actionable emulation guidance, but it is monolithic with no progressive disclosure and lacks validation checkpoints in its destructive/batch workflows.

Suggestions

Add explicit validation/verification steps to the emulation guidance (e.g., validate the side-loading triad executes in an isolated lab before propagation testing) so workflow clarity can exceed 3.

Split the sources list and the detailed TTP/campaign catalogs into reference files (e.g., references/sources.md, references/ttps.md) and link from SKILL.md to improve progressive disclosure and conciseness.

Tighten the notable-campaigns section to the most representative operations or move the full chronology to a reference, reducing token overhead while keeping the signature plays prominent.

DimensionReasoningScore

Conciseness

The body is dense, actor-specific threat intelligence (not concepts Claude already knows), but the 16-entry campaign list and 30-URL source list could be tightened or moved to references, fitting 'Mostly efficient but includes some unnecessary explanation or could be tightened'.

3 / 5

Actionability

The 'Emulation guidance' section gives concrete, executable direction with specific commands and parameters (ping -n 70, rar.exe a -hp, curl to Dropbox, TLS headers '17 03 03') and named tools (Sliver, Cobalt Strike, Mimikatz), with only minor gaps.

4 / 5

Workflow Clarity

Plays are organized as a categorized menu with implicit sub-sequences (the side-loading triad's steps 1-3) but lack explicit validation checkpoints or feedback loops for destructive operations (USB worm, kernel-mode driver), which caps workflow clarity at 3.

3 / 5

Progressive Disclosure

Section headers are clear and well-organized, but the skill is a monolithic 250+ line file with no bundle references; content such as the sources list and full TTP catalog that could live in separate reference files is inlined.

3 / 5

Total

13

/

20

Passed

Description

60%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 highly specific and distinctive, leveraging multiple actor aliases as strong trigger terms, but it omits concrete capability actions and any explicit 'Use when' guidance, capping specificity and completeness.

Suggestions

Add concrete actions to the description (e.g., 'emulate DLL side-loading chains, themed spearphishing lures, and USB propagation') to lift specificity above 2.

Append an explicit 'Use when ...' clause naming the trigger scenarios (e.g., 'Use when emulating Mustang Panda / Bronze President / TA416 campaigns') so completeness can exceed 3.

Surface additional natural aliases currently only in metadata (PlugX, Earth Preta, Camaro Dragon, HIVE0154) into the description to push trigger-term coverage toward 5.

DimensionReasoningScore

Specificity

Names the domain ('Adversary-emulation profile') and actor identity but lists no concrete actions such as emulating DLL side-loading or crafting lures, matching 'Names the domain but actions are minimal or generic'.

2 / 5

Completeness

Provides a clear 'what' (an adversary-emulation profile of the actor) but includes no 'Use when...' clause or explicit trigger guidance, so completeness caps at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

Includes six natural actor aliases ('Mustang Panda', 'G0129', 'Bronze President', 'Stately Taurus', 'RedDelta', 'TA416', 'TEMP.Hex') plus 'China-based state sponsored cyber-espionage', giving good keyword coverage, though additional synonyms (PlugX, Earth Preta, Camaro Dragon) live only in metadata.

4 / 5

Distinctiveness Conflict Risk

Targets a single named threat actor with multiple distinct aliases and a clear espionage/emulation niche, yielding minimal conflict risk with other skills.

5 / 5

Total

14

/

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/shared/adversary-emulation/mustang-panda/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.