CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

c2-havoc

Havoc C2 framework (C5pider/Havoc) — modern Sliver/CS alternative, Demon agent with indirect syscalls, sleep obfuscation (Ekko/Zilean/FOLIAGE), Donut PIC loader integration, profile-driven HTTP comms, MaterialUI web client. Best when you need modern OPSEC without Cobalt Strike cost.

Quality

73%

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/standard/post-exploit/c2/havoc/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

61%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.

A well-organized, technically credible operator reference with concrete build steps, a realistic profile example, and useful decision guidance. Its weaknesses are the absence of any validation checkpoints in a workflow that includes destructive operations (capping workflow clarity), plus stale time-dated notes, a dead code block, and general EDR explanation that pad the token budget.

Suggestions

Add validation checkpoints to the workflow: after starting the teamserver, verify the listener is active (e.g., '> listeners') before building the Demon, and confirm implant callback (e.g., '> demons') before running destructive or lateral-movement commands — this lifts workflow clarity past the destructive-operation cap.

Trim token waste: delete the empty CLI-build code block, move the time-sensitive 'as of 2024' / 'released 2023' notes into a clearly dated section, and compress the general direct-vs-indirect syscall primer to the Havoc-specific behavior that Claude would not already know.

Split the growing reference material — the comparison table, the sleep-obfuscation internals, and the common-commands list — into a references/ file (e.g., 'See [COMMANDS.md](references/commands.md)'), keeping SKILL.md as a lean overview with one-level-deep pointers.

DimensionReasoningScore

Conciseness

The body is mostly efficient — the profile config, build steps, and comparison table each earn their tokens — but it includes unnecessary or stale material: the empty code block '# CLI build (haven't found stable CLI build API as of 2024...)', time-sensitive dates ('released 2023', 'as of 2024') that the guidelines penalize outside a deprecation section, and a general direct-vs-indirect syscall EDR primer that largely restates knowledge Claude already has. This matches anchor 3 ('mostly efficient but includes some unnecessary explanation or could be tightened'), above anchor 2's multiple padded sections but below anchor 4's only minor trims.

3 / 5

Actionability

Mostly executable guidance: copy-paste build commands ('git clone ... make ts-build'), a complete yaotl profile, the server launch line, and concrete Demon UI build steps with named options. It falls short of anchor 5 because the payload build path cannot be executed end-to-end (UI-driven with an explicitly empty CLI block) and the common-commands section lists command forms like 'jump <psexec64|wmi|smb_lateral> <host>' without runnable examples — anchor 4's 'concrete code or commands with minor gaps'.

4 / 5

Workflow Clarity

The sequence (setup → configure profile → build Demon → operate) is present and ordered, but there are no validation or verification checkpoints anywhere (e.g., confirming the listener is live before building the Demon, or checking implant callback before running post-exploitation commands). Because the commands include destructive and batch operations ('proc kill <pid>', 'jump psexec64', lateral movement), the feedback-loop cap applies and workflow_clarity cannot exceed 3, which also fits anchor 3 ('sequence present but checkpoints missing or implicit').

3 / 5

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ are absent), and the body is organized into clear topical sections with a clearly signaled one-level References list of external URLs. It matches anchor 4 ('good structure; most content appropriately placed; minor organization gaps') rather than anchor 5 because at ~145 lines — beyond the under-50-line simple-skill case — content such as the OPSEC internals deep-dive, the comparison table, and the command reference could be split into reference files to keep SKILL.md a leaner overview.

4 / 5

Total

14

/

20

Passed

Description

73%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 strong, information-dense description that identifies the framework precisely and enumerates its distinguishing capabilities with correct technical naming. Its main gaps are the second-person 'when' clause (which also costs it the specificity penalty) and a trigger list that misses several natural phrasings users would employ.

DimensionReasoningScore

Specificity

The description lists several concrete capabilities — 'Demon agent with indirect syscalls, sleep obfuscation (Ekko/Zilean/FOLIAGE), Donut PIC loader integration, profile-driven HTTP comms' — which sits between anchor 3 (1-2 concrete actions) and anchor 4 (several specific actions); however, these are framework attributes rather than actions, and the second-person phrasing 'Best when you need modern OPSEC' triggers the mandated 1-point specificity reduction from 4 to 3.

3 / 5

Completeness

Both 'what' (a C2 framework with the listed OPSEC capabilities) and 'when' ('Best when you need modern OPSEC without Cobalt Strike cost') are present, but the 'when' clause covers only a single scenario and lacks concrete trigger phrasing like 'Use when the user mentions Havoc, Demon, or C2' — exactly anchor 4's 'when could be more explicit or specific', not anchor 5's explicit trigger phrases, and clearly above anchor 3 where 'when' is only implied.

4 / 5

Trigger Term Quality

Good keyword coverage with the natural terms an operator would say: 'Havoc', 'C2', 'Demon', 'sleep obfuscation', 'indirect syscalls', 'Ekko/Zilean/FOLIAGE', 'Donut', 'OPSEC', plus the synonym framing 'Sliver/CS alternative'. It falls short of anchor 5 because common phrasings like 'command and control', 'implant', 'red team', or 'post-exploitation' are absent, matching anchor 4 ('good keyword coverage; a few natural terms missing') better than anchor 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

It carves out a clear niche — the Havoc framework and its Demon agent specifically — with distinct trigger terms that no sibling skill (Sliver, Mythic, Cobalt Strike) would claim, matching anchor 5 ('clear niche with distinct triggers; minimal conflict risk') rather than anchor 4, which requires some overlap risk with closely related skills.

5 / 5

Total

16

/

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/standard/post-exploit/c2/havoc/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.