Prompt Injection PatternPublished Apr 12, 2026

Hidden instructions in SKILL.md YAML frontmatter description field

HighAARM tierConfirmed
Seen212times across22customers
jiffy-ti-2026-000010

Summary

Attacker exploits the fact that agents often read the full description field verbatim before deciding whether to invoke a skill. Description contains multi-line instructions masquerading as usage notes but actually directing the agent to read .env or install a second skill.

Severity & confidence

Severity
High

High

Credible abuse path with meaningful impact on confidentiality, integrity, or availability. Remediate on a short timeline; do not wait for the next sprint cycle.

Confidence
Confirmed

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.

  • description\s*:\s*[^\n]{200,}
  • description[\s\S]{0,500}(ignore|disregard|new instructions)

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_000010
{
    meta:
        source = "jiffy-intel"
        severity = "high"
        description = "Auto-generated from Jiffy Intel indicators"
    strings:
    $content_pattern_0 = "description\\s*:\\s*[^\\n]{200,}"
    $content_pattern_1 = "description[\\s\\S]{0,500}(ignore|disregard|new instructions)"
    condition:
        $content_pattern_0 or $content_pattern_1
}

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

Affected tools

ToolVersionsStatus
Claude Code*vulnerable
Any skill-loading agent*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.

  • summary-writer-skillSkill
    Removed
    Source
    Anthropic Skills
    Versions
    1.0.0
    First observed
    Mar 13, 2026
    Last observed
    Apr 3, 2026

    SKILL.md description field stuffed with prompt-override instructions between benign sentences.

  • agents-md-reviewerIDE rules
    Under review
    Source
    GitHub (public)
    First observed
    Mar 20, 2026

    agents.md pattern — same hidden-instruction technique applied to the newer spec.

  • changelog-generator-skillSkill
    Quarantined
    Source
    Community registry
    Versions
    0.7.x
    First observed
    Mar 27, 2026

How to remediate

  1. 01Cap SKILL.md description fields at 300 chars.
  2. 02Flag descriptions containing instruction-override phrases.
  3. 03Jiffy Layer 3 semantic analysis scores these at intent_confidence >= 0.7.

Timeline & sources

Timeline

  1. First observedApr 12, 20265 days ago
  2. Last updatedApr 12, 20265 days ago
  3. PublishedApr 12, 20265 days ago

Sources

scannercurated

References

Scan for patterns like this

Point Jiffy at your GitHub org, IDE config, or a single artifact. Get a scored report in under a minute.

Start a free scan