Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Compliance Image Guard — measured AI agent skill | DecimalAI

stijnman/compliance-image-guard

SkillScore85Maintained

Checks images for policy compliance before generation or publish. Use before sharing images or when user says compliance check, safe image. Use when the user needs this capability. Triggers: compliance check, safe image, image p olicy,can I publish this.

by stijnman

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

AI Engineer pack

decimalai skills pull stijnman-compliance-image-guard

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/stijnman-compliance-image-guard/SKILL.md

Raw file: /s/stijnman-compliance-image-guard/SKILL.md ↗ · machine-readable summary at /s/stijnman-compliance-image-guard.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

26 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 7d ago

Yes — +27% lift-34% tokens0% turns85% of whole cases pass with skill (22/26)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	-7%	0%
case-02	✗→✓	▲ Improved	-2%	0%
case-03	✗→✓	▲ Improved	-31%	0%
case-11	✗→✓	▲ Improved	-47%	0%
case-12	✗→✓	▲ Improved	-32%	0%

Showing 5 of 26 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/1/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

stijnman / grok-custom-skills

.grok/skills/compliance-image-guard/SKILL.md

View on GitHub ↗

License MITImported 8/1/2026Version v1

Skill contents

SKILL.md

Compliance Image Guard

When to Use

User says compliance check or task matches this capability
User says safe image or task matches this capability
User says image policy or task matches this capability
User says can I publish this or task matches this capability

Workflow

Review image prompt or asset for policy risks.
Flag: real persons without consent, violence, IP infringement.
Block or revise prompt; route edge cases to hitl-approver.
Log decision rationale.

Integrations

safe-image-editor
imagine-asset-generator
hitl-approver

Error Handling

| Failure | Response | |---------|----------| | Ambiguous policy | Default deny; ask user to confirm. |

Gotchas

Real-person likeness requires explicit user confirmation.

Example

Input: User request matching triggers above. Output: Structured result per workflow with integrations invoked as needed.

Want to see if stijnman/compliance-image-guard would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.