Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/kernora-ai/agent-sec/agentsec-reportgit clone --depth 1 https://github.com/kernora-ai/agent-secWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/kernora-ai/agent-sec/agentsec-report)<a href="https://agentmods.dev/commands/kernora-ai/agent-sec/agentsec-report"><img src="https://agentmods.dev/badge/commands/kernora-ai/agent-sec/agentsec-report.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00031 | $0.00365 |
| Opus 5 | $0.00015 | $0.00182 |
| Sonnet 5 | $0.00006 | $0.00073 |
| Haiku 4.5 | $0.00003 | $0.00036 |
Grade A, and why
agentsec-report scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Read the LOCAL advisory log and summarize it for the user. Run:
!python3 - <<'PY' import json,os,collections,time p=os.path.expanduser("~/.agentsec/flags.log") if not os.path.exists(p): print("No flags yet — the advisory hook hasn't caught anything in your sessions (or isn't enabled)."); raise SystemExit rows=[json.loads(l) for l in open(p) if l.strip()] if not rows: print("No flags recorded yet."); raise SystemExit total=len(rows); rules=sum(r.get("rules",0) for r in rows) first=time.strftime("%Y-%m-%d",time.localtime(min(r["t"] for r in rows))) print(f"Kernora Agent Security — your local effectiveness view (nothing left your machine):") print(f" {total} risky action(s) flagged since {first}; {rules} rule-citation(s) surfaced.") c=collections.Counter(r["cmd"].split()[0] if r.get("cmd") else "?" for r in rows) print(" most-flagged command starts: " + ", ".join(f"{k}×{v}" for k,v in c.most_common(5))) PY
Then explain: this is a LOCAL view only (we store nothing server-side — the free tier is stateless and private). Full per-action reporting with a tamper-evident, cited audit ledger is the paid Kernora Axiora Integrity Plane.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 24 lines · 31 tokens per session scan A 2f46333e4f02
agentsec-report is a command published in the GitHub repository kernora-ai/agent-sec (1 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 31 tokens to every session and 365 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
notify-authors
Open GitHub issues on flagged repos to notify authors before public disclosure.
scan-daily
Run incremental ecosystem scan: crawl registries → audit new packages → merge results → generate posts.
scan-skill
Scan a specific MCP skill by npm package name or GitHub URL.
wf-ads
Generate 50-100 ad copy variants from pain point research, then create renderable ad images.
wf-research
Research real customer pain points from Reddit, forums, and social media to fuel ad copy and content.
launch-checklist
Run through the complete launch readiness checklist for Panguard AI.