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 instructions/vulnerscom/api/agents-mdgit clone --depth 1 https://github.com/vulnersCom/apiWhat 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 | $0.01497 | $0.01497 |
| Opus 5 | $0.00749 | $0.00749 |
| Sonnet 5 | $0.00299 | $0.00299 |
| Haiku 4.5 | $0.00150 | $0.00150 |
Grade A, and why
api AGENTS.md 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — using the Vulners Python SDK
Guidance for AI agents (and their humans) using this SDK to work with the
Vulners vulnerability intelligence API. For contributor/development
conventions, see .agents/skills/vulners-api/SKILL.md instead.
Install & authenticate
pip install vulners # core
pip install "vulners[mcp]" # + the MCP server (see below)
Working inside a repository checkout? Use
uv sync --lockedanduv run python .... Do not runpip install vulners— that installs the currently published release rather than this checkout. (For development conventions and commands, see.agents/skills/vulners-api/SKILL.md.)
The API key comes from the VULNERS_API_KEY environment variable, or is passed explicitly.
Get a free key at https://vulners.com. Never hard-code a key in committed code.
Two ways to use it
1. As a library
Prefer the v4 clients: Vulners (sync) and AsyncVulners (async). Both are context
managers and expose the same resource namespaces.
from vulners import Vulners
with Vulners() as v: # reads VULNERS_API_KEY
# Search (Lucene syntax → typed Bulletin models; access fields as attributes)
for b in v.search.query("type:cve AND cvss.score:[9 TO 10]", limit=10):
print(b.id, b.title, b.cvss and b.cvss.score)
# Single lookup (-> Bulletin | None)
cve = v.search.get_bulletin("CVE-2021-44228")
# Audit software / Linux host
v.audit.software(["cpe:2.3:a:apache:log4j:2.14.1"])
v.audit.linux_audit(os_name="debian", os_version="10",
packages=["openssl 1.1.1d-0+deb10u3 amd64"])
Key entry points:
| Namespace | What it does | Common methods |
|---|---|---|
v.search |
search & fetch documents | query, iter_query, get_bulletin, get_multiple_bulletins |
v.documents |
document-centric lookups | get, get_many, references, history |
v.audit |
vulnerability assessment | software, host, linux_audit, library_audit, sbom_audit, cve_audit, kb_audit, win_audit, smart |
v.archive |
bulk dataset download | fetch_collection, iter_collection (stream), fetch_collection_update |
v.misc |
lookups | search_cpe, query_autocomplete, get_suggestion |
v.subscriptions |
current v4 subscriptions API | list, get, create, update, delete |
v.subscriptions_email |
legacy v3 email subscriptions | list, add, edit, delete |
v.webhooks |
legacy v3 polling webhook subscriptions | list, add, read, delete |
v.report · v.stix · v.vscanner |
reporting, STIX bundles, VScanner | — |
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.
- yesterday First seen · 110 lines · 1,497 tokens per session scan A 0b4fd539783b
api AGENTS.md is an instructions file published in the GitHub repository vulnersCom/api (371 stars, last pushed 11d ago), licensed MIT. It adds 1,497 tokens to every session, about $0.0075 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 2026-08-30.
Other instructions, from other repositories
yandex-tracker-mcp AGENTS.md
Instructions for aikts/yandex-tracker-mcp, covering agents.md, project overview, commands, architecture and talking to the tracker api.
ErisPulse AGENTS.md
Instructions for ErisPulse/ErisPulse, covering erispulse 智能体开发事项, 代码修改, 测试与检查, ci/cd and 国际化(i18n).
atlan-python AGENTS.md
Instructions for atlanhq/atlan-python, covering security, contact, quickstart for agents, security invariants and data classification.
atlan-python CLAUDE.md
Instructions for atlanhq/atlan-python, covering security, contact, quickstart for agents, security invariants and review checklist.
OctoForge AGENTS.md
AGENTS.md instructions for dmirain/OctoForge, covering agents.md, the project, layout, commands and code conventions.
OctoForge CLAUDE.md
Claude Code instructions for dmirain/OctoForge, covering claude.md, commands, runtime reference and rules that bite.