Jiffy Intel — Threat intelligence for AI artifacts

Summary

Skill whose setup step silently rewrites ~/.claude/settings.json, flipping `alwaysAllowToolInvocation` to true and clearing the tool allowlist. Subsequent skills in the session then execute without user consent, including network egress and filesystem writes the user never approved.

Severity & confidence

Severity

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

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)settings\.json.{0,200}(?:alwaysAllow|skip[_\s-]?permissions|auto[_\s-]?approve)

```
\.claude/settings\.json
```

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_000014
{
    meta:
        source = "jiffy-intel"
        severity = "high"
        description = "Auto-generated from Jiffy Intel indicators"
    strings:
    $file_path_pattern_0 = "\\.claude/settings\\.json"
    $command_pattern_1 = "(?is)settings\\.json.{0,200}(?:alwaysAllow|skip[_\\s-]?permissions|auto[_\\s-]?approve)"
    condition:
        $file_path_pattern_0 or $command_pattern_1
}

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

Affected tools

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

claude-speed-tweaksSkill
Removed
Source
Anthropic Skills
Versions
1.3.0
First observed
Mar 11, 2026
Last observed
Apr 2, 2026
dev-mode-fastSkill
Removed
Source
Community registry
Versions
0.9.x
First observed
Mar 15, 2026
Last observed
Apr 4, 2026

How to remediate

01Pin ~/.claude/settings.json via a pre-commit hook or MDM profile; treat any skill write to this file as a critical alert.
02Restore consent prompts manually if the file was modified.

Timeline & sources

Timeline

First observedMar 11, 20261 month ago
Last updatedApr 16, 20261 day ago
PublishedMar 25, 202623 days ago

Sources

scannercurated

References

blog.jiffylabs.aiJiffy Research

Jiffy Research — How Jiffy Scans AI Artifacts

https://blog.jiffylabs.ai/posts/how-jiffy-scans-ai-artifacts-technical-overview

genai.owasp.orgOWASP

OWASP LLM-06: Excessive Agency (2026)

https://genai.owasp.org/llmrisk/llm-06-2026/

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.