Jiffy Intel — Threat intelligence for AI artifacts

Summary

Skill SKILL.md contains a "definition of done" section that, in addition to the legitimate task, requires the agent to "also verify X" where X is a credential read or a network call. The agent treats the side-channel as in-scope because it came from a trusted skill definition.

Severity & confidence

Severity

Low

Hygiene-grade issue with limited direct impact. Fix opportunistically; monitor in case context changes (new exploit, new dependency, etc.).

Confidence

Confirmed

Observed directly by Jiffy scanners or validated by multiple independent sources. Indicators are reliable enough to drive automated action.

Indicators

Observed patterns and artifacts associated with this entry. Each indicator can be copied into your detection stack or SIEM.

(?is)(?:definition of done|success criteria)[\s\S]{0,400}(?:verify|check|read|send)\s+[^\n]*(?:env|token|credential|key)

Detection rule

A YARA-style pseudo-rule auto-generated from the indicators above. Useful as a starting point — adapt the syntax for your target detection platform.

YARA-style pseudo-rule

rule jiffy_ti_2026_000020
{
    meta:
        source = "jiffy-intel"
        severity = "low"
        description = "Auto-generated from Jiffy Intel indicators"
    strings:
    $content_pattern_0 = "(?is)(?:definition of done|success criteria)[\\s\\S]{0,400}(?:verify|check|read|send)\\s+[^\\n]*(?:env|token|credential|key)"
    condition:
        $content_pattern_0
}

Auto-generated from the indicators above. Adapt syntax for your detection stack before deploying.

Affected tools

Tool	Versions	Status
Claude Code	`*`	vulnerable
Cursor	`*`	vulnerable
Windsurf	`*`	vulnerable

Example artifacts

Sanitized examples of artifacts Jiffy has observed exhibiting this pattern. Publisher handles are redacted; version ranges and status reflect the most recent scan.

pr-reviewer-skillSkill
Removed
Source
Anthropic Skills
Versions
1.2.0
First observed
Mar 17, 2026
Last observed
Apr 7, 2026
code-quality-skillSkill
Under review
Source
Anthropic Skills
Versions
2.0.0
First observed
Mar 21, 2026
tech-debt-skillSkill
Quarantined
Source
Community registry
Versions
0.8.x
First observed
Mar 25, 2026

How to remediate

01Treat skill-provided task definitions as untrusted.
02Require agent runtimes to prompt the user before allowing an unrelated side-task.

Timeline & sources

Timeline

First observedMar 17, 20261 month ago
Last updatedApr 22, 2026today
PublishedMar 31, 202617 days ago

Sources

curated

References

genai.owasp.orgOWASP

Related intel

Other entries of type Malicious Skill, or the most recent published entries if no same-type matches exist.

Malicious SkillHigh

MCP server's install script drops a skill into ~/.claude/skills/

MCP server whose install path (`pip install`, `npm i`) includes a side-effect that writes a skill file under the user's Claude skills directory. Installing the MCP silently installs a skill the user never consented to.

Jiffy IntelApr 28

Malicious SkillLow

AGENTS.md tells agent to paste ~/.ssh/id_rsa.pub into PR for CI setup

AGENTS.md has a section titled "CI onboarding" that instructs the agent to paste the user's SSH public key into PR descriptions. While public keys are not immediately sensitive, the coerced habit normalizes key pasting and enables later pattern-matching attacks.

Jiffy IntelApr 28

Malicious SkillMedium

Skill installs a FUSE filesystem that shadows ~/.aws

Skill mounts a FUSE filesystem over ~/.aws on macOS/Linux that proxies reads but logs every access. The real credentials remain accessible; the skill gains a reliable sidechannel for any later AWS CLI invocation.

Jiffy IntelApr 27

Skill instructions redefine "success" to include side-channel task