agentsclimarketplace

Agent DLP Guard

Skill axtontc/Agent-DLP-Guard

Data loss prevention guardrail for autonomous agents.

Install
npx -y skills add axtontc/Agent-DLP-Guard

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 26 days oldThe repository was created 26 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

**STOP AND VERIFY**: Before running any command or tool that results in irreversible data loss, you MUST obtain explicit user consent. When in doubt, ask. It is better to wait for confirmation than to accidentally delete production data or critical project assets.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

2.3 KB, 467 tokens by cl100k_base, as published. Nobody here has run it

Agent-DLP: Zero-Trust Guardrail Protocol

[!CAUTION]

STOP AND VERIFY: Before running any command or tool that results in irreversible data loss, you MUST obtain explicit user consent.

🛡️ Scope of Interception

If your planned actions include ANY of the following patterns, you must halt execution and engage the mandatory procedure below:

1. File System Destruction

  • Recursive force deletes: rm -rf, del /s /q
  • Overwriting critical system configs (/etc/, ~/.bashrc, C:\Windows\)
  • Emptying entire directories containing source code.

2. Database Dropping

  • SQL commands containing: DROP TABLE, DROP DATABASE, DROP SCHEMA.
  • TRUNCATE TABLE operations.
  • Broad DELETE FROM statements missing a strict WHERE clause or utilizing 1=1.

3. Cloud Infrastructure Demolition

  • Terraform: terraform destroy
  • AWS: aws s3 rb --force, aws ec2 terminate-instances
  • GCP: gcloud projects delete, bq rm -r -f, gcloud storage rm
  • Key Management: Deleting secrets or scheduling KMS key destruction.

4. Source Control Sabotage

  • Git force pushes to protected branches: git push --force origin main
  • Hard resets that drop unpublished commits: git reset --hard
  • Remote branch deletion.

🚦 Mandatory Procedure

  1. Halt Execution: Do not execute the command, script, or API call.
  2. Request Consent: Output a clear, human-readable message explaining:
    • The impact of this deletion (e.g., "This will permanently delete the production database").
    • Why you believe this action is necessary based on the current context.
    • A request for the user's explicit, affirmative approval to proceed.
  3. Wait: End your turn. Only proceed if the user responds with clear consent. If the user denies or questions the action, abort and suggest a non-destructive alternative (e.g., soft-deletes, renaming, or backing up first).

What ships with it: 4 files

7.6 KB alongside SKILL.md, 1 of them executable

examples/

Keep looking

Skills are one crate of 326,984. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.