All loops
SecurityHard 15 min· claude-opus-4-1
SBOM diff narrator
SBOM diff narrator with audit-grade logs.
Read the full Security build guide →NOT DEPLOYEDNOT DEPLOYED
0156ms
Trigger
cron(0 7 * * *) fired · every day · 07:00
021056ms
Agent
claude-opus-4-1 · in 1356 tok · out 412 tok
03326ms
Tools
aws-mcp/anthropic:messages.create → 200 OK · 296ms
0446ms
Verify
schema check · json-schema draft-2020 passed
0576ms
Output
linear ticket · SOC2 evidence stored
0636ms
Notify
audit log written · runbook link attached
SUCCESS
0%
0 runs
P50
0ms
median
P95
0ms
tail
AVG COST
—
per run
LAST OK
never
—
LAST FAIL
never
none
Latency · last 30 runs0 samples
no runs yet
Latest output · what your users see
SEC-1076CWE-798
Outdated `axios@0.21.1` — CVE-2023-45857 (SSRF)
// press Test to run once · Watch live to keep streaming · Deploy to make it real
The problem
Security tasks are tedious, easy to skip, and dangerous to skip.
The outcome
Tedium is automated; humans review summaries and approve.
Ingredients & skills
Secrets
- ANTHROPIC_API_KEY
Providers
- Anthropic
MCP servers
- aws-mcp
- github-mcp
#security#audit#claude
How it works
SBOM diff narrator with audit-grade logs.
Step 1
1 — Read-only scan
All cloud + repo calls are scoped read-only by the locker.
Step 2
2 — Summarise
Claude ranks findings by blast radius, not raw severity.
Step 3
3 — Ticket
Top findings open tickets with owner + suggested fix.
One-line deploy
The button above runs the same command with your saved config. This is the raw CLI form.
bash
npx claudeloops deploy sbom-diff-narrator