clawdstrike 0.2.5

Security guards and policy engine for AI agent execution
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

# Permissive Ruleset
# Development-friendly: allows all egress and relaxes patch limits.
version: "1.1.0"
name: Permissive
description: Permissive rules for development (use with caution)

guards:
  egress_allowlist:
    allow:
      - "*"
    block: []
    default_action: allow

  patch_integrity:
    max_additions: 10000
    max_deletions: 5000
    require_balance: false
    max_imbalance_ratio: 50.0

settings:
  fail_fast: false
  verbose_logging: true
  session_timeout_secs: 7200