ink5725/snapshot-config-reader agent skill scanner finding
A skill that extracts config from PNG images triggered scanner flags. The snapshot showed zero downloads and an incomplete scan; the finding does not establish that the skill is malicious.
What happened
Agent skill SKILL.md instructs pulling 'config' from PNG trailer, EXIF, LSB. ClamAV Img.Packed.PngContainsDownloadCmd (high) on two PNGs; Nova DetectHiddenInstructions (low). 0 downloads; scan incomplete.
How it happened
Read the flow as text & sources
A catalog reconstruction of the supplied source account. Scanner finding on a skill bundle; Jose explicitly not calling it malicious. Pattern of decode-from-image instructions.
- 01 · Skill bundle is scanned
The finding concerns an agent skill with no observed downloads.
Hunt: A skill bundle introduces new network, execution or secret-access instructions, including instructions distributed across non-Markdown files.
Scanner findings and evasion results are research evidence, not a runtime compromise detector. Do not execute embedded commands or invent a victim chain.
Source for step 1 - 02 · SKILL.md requests image extraction
Instructions tell the agent to obtain configuration from PNG trailers, metadata or low bits.
Hunt: A skill bundle introduces new network, execution or secret-access instructions, including instructions distributed across non-Markdown files.
Scanner findings and evasion results are research evidence, not a runtime compromise detector. Do not execute embedded commands or invent a victim chain.
Source for step 2 - 03 · Scanners flag embedded instructions
The partial scan detects suspicious image content with differing confidence levels.
Hunt: A skill bundle introduces new network, execution or secret-access instructions, including instructions distributed across non-Markdown files.
Scanner findings and evasion results are research evidence, not a runtime compromise detector. Do not execute embedded commands or invent a victim chain.
Source for step 3 - 04 · No execution is established
The record does not confirm a victim, a completed scan or successful skill execution.
Source for step 4
Why it matters
Reported entry point: n/a (skill install / agent load path). Reported impact: 0 downloads at time of post.
Check these source-linked conditions against your setup.
The scanner finding concerns an agent skill that retrieves instructions from image content.
Conditions to check- Confirm whether this specific bundle or the described extraction pattern is present.
- The scan was incomplete, the snapshot reported zero downloads and maliciousness was not established.
Source-linked applicability guidance. Confirm current scope in the original report.
What you can do
Protection & prevention
Review the full skill bundle and decoded-image instructions before approving its use.
Whole-bundle skill scanning; treat decode-from-image steps as signals
Catalog defensive guidance based on the reported behavior. Confirm scope and current mitigations in the original source.
Detection queries
Draft query starters grouped by the attack stages where they apply.
Start with the logs you collect. Each hunt identifies its required visibility, field mapping and limits. Queries are catalog-authored drafts using reusable behavior profiles; test with known activity and your normal baseline before alerting.
01Skill bundle is scannedSkill bundle provenance / reviewCustom schema · Draft
The finding concerns an agent skill with no observed downloads.
What to look for
A skill bundle introduces new network, execution or secret-access instructions, including instructions distributed across non-Markdown files.
Visibility: Review / research workflow
Required logs
- Complete skill-bundle inventory and hashes
- Scanner results, extracted attachments and review history
- Agent load / tool logs if execution is actually observed
Coverage limit. Scanner findings and evasion results are research evidence, not a runtime compromise detector. Do not execute embedded commands or invent a victim chain.
Custom-schema candidates, not out-of-box detectors. Map your application / cloud / platform logs to EventType, Actor, Asset, Action, Target and CorrelationId. KQL uses a custom OpenAttacksAudit_CL table; SPL uses openattacks:audit. EventType values are proposed normalization names, not native vendor event names. Correlation and environment-specific alert logic remain to be built.
Sigma
title: Hunt - Skill bundle provenance / review
id: 4941aba2-b392-4f04-ad39-6f0201811d26
status: experimental
description: A skill bundle introduces new network, execution or secret-access
instructions, including instructions distributed across non-Markdown files.
author: MagicSword OpenAttacks catalog
date: 2026-10-23
falsepositives:
- Authorized activity matching this broad hunt; baseline and investigate
before alerting.
level: low
logsource:
category: application
product: openattacks_audit
definition: Custom normalized audit schema; map EventType, Actor, Asset, Action,
Target and CorrelationId from the required source logs first.
detection:
selection:
EventType:
- skill.bundle.load
- skill.bundle.change
condition: selection
KQL
// SCHEMA TEMPLATE: create/map this custom table before use; not a built-in Sentinel table.
// OpenAttacksAudit_CL: TimeGenerated(datetime), EventType/Actor/Asset/Action/Target/CorrelationId(string).
// Selects candidates only. Apply the stage hypothesis and correlation steps before alerting.
OpenAttacksAudit_CL
| where TimeGenerated > ago(7d)
| where EventType in ("skill.bundle.load", "skill.bundle.change")
| project TimeGenerated, EventType, Actor, Asset, Action, Target, CorrelationId
| order by TimeGenerated descSPL
sourcetype="openattacks:audit"
EventType IN ("skill.bundle.load", "skill.bundle.change")
| table _time EventType Actor Asset Action Target CorrelationId
| sort - _timeInvestigation steps
- Review the entire bundle, including images and referenced files, in an isolated static-analysis workflow.
- Compare requested capabilities with the skill's purpose; record scanner disagreement and avoid treating a finding as proof of execution.
02SKILL.md requests image extractionSkill bundle provenance / reviewCustom schema · Draft
Instructions tell the agent to obtain configuration from PNG trailers, metadata or low bits.
What to look for
A skill bundle introduces new network, execution or secret-access instructions, including instructions distributed across non-Markdown files.
Visibility: Review / research workflow
Required logs
- Complete skill-bundle inventory and hashes
- Scanner results, extracted attachments and review history
- Agent load / tool logs if execution is actually observed
Coverage limit. Scanner findings and evasion results are research evidence, not a runtime compromise detector. Do not execute embedded commands or invent a victim chain.
Custom-schema candidates, not out-of-box detectors. Map your application / cloud / platform logs to EventType, Actor, Asset, Action, Target and CorrelationId. KQL uses a custom OpenAttacksAudit_CL table; SPL uses openattacks:audit. EventType values are proposed normalization names, not native vendor event names. Correlation and environment-specific alert logic remain to be built.
Sigma
title: Hunt - Skill bundle provenance / review
id: 4941aba2-b392-4f04-ad39-6f0201811d26
status: experimental
description: A skill bundle introduces new network, execution or secret-access
instructions, including instructions distributed across non-Markdown files.
author: MagicSword OpenAttacks catalog
date: 2026-10-23
falsepositives:
- Authorized activity matching this broad hunt; baseline and investigate
before alerting.
level: low
logsource:
category: application
product: openattacks_audit
definition: Custom normalized audit schema; map EventType, Actor, Asset, Action,
Target and CorrelationId from the required source logs first.
detection:
selection:
EventType:
- skill.bundle.load
- skill.bundle.change
condition: selection
KQL
// SCHEMA TEMPLATE: create/map this custom table before use; not a built-in Sentinel table.
// OpenAttacksAudit_CL: TimeGenerated(datetime), EventType/Actor/Asset/Action/Target/CorrelationId(string).
// Selects candidates only. Apply the stage hypothesis and correlation steps before alerting.
OpenAttacksAudit_CL
| where TimeGenerated > ago(7d)
| where EventType in ("skill.bundle.load", "skill.bundle.change")
| project TimeGenerated, EventType, Actor, Asset, Action, Target, CorrelationId
| order by TimeGenerated descSPL
sourcetype="openattacks:audit"
EventType IN ("skill.bundle.load", "skill.bundle.change")
| table _time EventType Actor Asset Action Target CorrelationId
| sort - _timeInvestigation steps
- Review the entire bundle, including images and referenced files, in an isolated static-analysis workflow.
- Compare requested capabilities with the skill's purpose; record scanner disagreement and avoid treating a finding as proof of execution.
03Scanners flag embedded instructionsSkill bundle provenance / reviewCustom schema · Draft
The partial scan detects suspicious image content with differing confidence levels.
What to look for
A skill bundle introduces new network, execution or secret-access instructions, including instructions distributed across non-Markdown files.
Visibility: Review / research workflow
Required logs
- Complete skill-bundle inventory and hashes
- Scanner results, extracted attachments and review history
- Agent load / tool logs if execution is actually observed
Coverage limit. Scanner findings and evasion results are research evidence, not a runtime compromise detector. Do not execute embedded commands or invent a victim chain.
Custom-schema candidates, not out-of-box detectors. Map your application / cloud / platform logs to EventType, Actor, Asset, Action, Target and CorrelationId. KQL uses a custom OpenAttacksAudit_CL table; SPL uses openattacks:audit. EventType values are proposed normalization names, not native vendor event names. Correlation and environment-specific alert logic remain to be built.
Sigma
title: Hunt - Skill bundle provenance / review
id: 4941aba2-b392-4f04-ad39-6f0201811d26
status: experimental
description: A skill bundle introduces new network, execution or secret-access
instructions, including instructions distributed across non-Markdown files.
author: MagicSword OpenAttacks catalog
date: 2026-10-23
falsepositives:
- Authorized activity matching this broad hunt; baseline and investigate
before alerting.
level: low
logsource:
category: application
product: openattacks_audit
definition: Custom normalized audit schema; map EventType, Actor, Asset, Action,
Target and CorrelationId from the required source logs first.
detection:
selection:
EventType:
- skill.bundle.load
- skill.bundle.change
condition: selection
KQL
// SCHEMA TEMPLATE: create/map this custom table before use; not a built-in Sentinel table.
// OpenAttacksAudit_CL: TimeGenerated(datetime), EventType/Actor/Asset/Action/Target/CorrelationId(string).
// Selects candidates only. Apply the stage hypothesis and correlation steps before alerting.
OpenAttacksAudit_CL
| where TimeGenerated > ago(7d)
| where EventType in ("skill.bundle.load", "skill.bundle.change")
| project TimeGenerated, EventType, Actor, Asset, Action, Target, CorrelationId
| order by TimeGenerated descSPL
sourcetype="openattacks:audit"
EventType IN ("skill.bundle.load", "skill.bundle.change")
| table _time EventType Actor Asset Action Target CorrelationId
| sort - _timeInvestigation steps
- Review the entire bundle, including images and referenced files, in an isolated static-analysis workflow.
- Compare requested capabilities with the skill's purpose; record scanner disagreement and avoid treating a finding as proof of execution.
Technical detailsTooling, impact & MITRE ATLAS
Attack context
- AI's role
- Agent / MCP as target
- Initial access
- n/a (skill install / agent load path)
Tooling as reported
AI products, models and artifacts
- Agent skill bundle
Other tooling
- PNG steganography / trailer / EXIF / LSB
Targets and reported impact
- Reported scale / exposure
- 0 downloads at time of post
Detection rules
- Review attack-specific indicatorsGUIDANCE
Monitor for the techniques described in ink5725/snapshot-config-reader skill finding. Refer to sources for specific IOCs and TTPs.
Source defensive notes
Whole-bundle skill scanning; treat decode-from-image steps as signals
MITRE ATLAS
AI-specific catalog annotations based on the reported behavior; these are not MITRE case-study assignments.
The flagged skill instructs an agent to recover instructions from image artifacts; actual execution is unconfirmed. Source
Sources & research credit
1 sourceResearch by Jose Enrique Hernandez (LOLSkills feed) · MagicSword / LOLSkills
Published analysis: MagicSword
Record & downloads
- Source record name
- ink5725/snapshot-config-reader skill finding
- Entry ID
- lolskills-ink5725-snapshot-config-reader-2026-10
- Catalog review
- October 20, 2026
- Category
- attacks-on-ai
- AI involvement confidence
- unverified