Jiffy Intel — Threat intelligence for AI artifacts

Summary

agents.md file that looks like ordinary prose but uses zero-width space characters to embed hidden tokens. The raw file, read by the agent, carries directives the human reviewer cannot see.

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.

```
[\u200B-\u200D\uFEFF]{3,}
```

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_000064
{
    meta:
        source = "jiffy-intel"
        severity = "low"
        description = "Auto-generated from Jiffy Intel indicators"
    strings:
    $content_pattern_0 = "[\\u200B-\\u200D\\uFEFF]{3,}"
    condition:
        $content_pattern_0
}

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

Affected tools

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

agents.md (repo: onboarding-kit)IDE rules
Under review
Source
GitHub (public repo)
First observed
Mar 13, 2026
.cursorrules (repo: team-conventions)IDE rules
Quarantined
Source
GitHub (public repo)
First observed
Mar 17, 2026

How to remediate

01Strip zero-width characters from IDE rule files before feeding them to the agent.
02Jiffy Layer 1 flags ZWSP runs as suspicious.

Timeline & sources

Timeline

First observedMar 13, 20261 month ago
Last updatedApr 26, 2026today
PublishedMar 25, 202623 days ago

Sources

scanner

References

genai.owasp.orgOWASP

OWASP LLM-01: Prompt Injection (2026)

https://genai.owasp.org/llmrisk/llm-01-2026/

blog.jiffylabs.aiJiffy Research

Jiffy Research — .cursorrules and agents.md Config Backdoors

https://blog.jiffylabs.ai/posts/cursorrules-and-agents-md-config-backdoors

Related intel

Other entries of type Prompt Injection Pattern, or the most recent published entries if no same-type matches exist.

Prompt Injection PatternHigh

agents.md writes to CLAUDE.md at runtime

agents.md tells the agent to "update CLAUDE.md with learnings from this session." The agent, executing the directive, writes attacker-crafted text into the pinned CLAUDE.md, poisoning future sessions.

Jiffy IntelApr 28

Prompt Injection PatternMedium

Claude Project instructions persist across team members' sessions

Shared projects carry instructions into every team member's sessions. A compromised project owner can silently push a malicious directive that affects all downstream usage — effectively a persistent cross-user prompt injection.

Jiffy IntelApr 28

OtherLow

.cursorrules contains large wall-of-text that pushes user intent out of context

.cursorrules whose rule body exceeds 20 000 characters of filler content. Each agent invocation consumes the rule, leaving limited context for the user prompt. Not a direct exfiltration primitive — a budget-starvation attack.

Jiffy IntelApr 28

agents.md uses zero-width whitespace to hide directives in innocuous text

Summary

Severity & confidence

Low

Confirmed

Indicators

Detection rule

Affected tools

Example artifacts

How to remediate

Timeline & sources

Timeline

Sources

References

OWASP LLM-01: Prompt Injection (2026)

Jiffy Research — .cursorrules and agents.md Config Backdoors

Scan for patterns like this