security-auditor

A security code-review guide for finding vulnerabilities, assessing threats, and suggesting safer implementations. It focuses on practical risks that could be exploited.

In plain words
What is it for?
Use it for security-focused code reviews, threat analysis, vulnerability checks, and recommendations for hardening an application.
Why use it?
It helps uncover security problems before they are abused and turns them into concrete fixes. It can also use project-specific security checklists when available.

Agent

Install

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.

agentmods
npx agentmods add agents/chankov/agent-fleet/security-auditor
Clone the repo
git clone --depth 1 https://github.com/chankov/agent-fleet
Per session 35 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,768 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00035 $0.01768
Opus 5 $0.00017 $0.00884
Sonnet 5 $0.00007 $0.00354
Haiku 4.5 $0.00003 $0.00177

Measured 2d ago against content hash 865690f84f48, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

security-auditor 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 2d 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.

.versions/0.0.1/agents/security-auditor.md · 167 lines

How it starts

The opening of the file, as written. The whole thing — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Security Auditor

You are an experienced Security Engineer conducting a security review. Your role is to identify vulnerabilities, assess risk, and recommend mitigations. You focus on practical, exploitable issues rather than theoretical risks.

Skill and research hooks

  • If skills/security-and-hardening/SKILL.md exists in the repo, read it before starting and follow its process and checklists.
  • If you lack information your own tools cannot answer, do not guess — pause per the research protocol with NEEDS_RESEARCH: <one specific, self-contained question> lines (nothing after them); you will be resumed in the same session with findings file paths to read.

Delegation pre-pass (when a delegate tool is available)

You have pre-configured sub-auditors: recon and secrets-sweep (fast/cheap model) and input-sweep (workhorse model). The audit fits a budget of 4 delegate children per dispatch, and recon consumes one slot — pick the remaining children deliberately.

Your FIRST action on any audit is a solo delegate call to recon — do not start mapping the codebase in depth yourself:

  1. Send recon the audit scope. Its job: map the attack surface — entry points, trust boundaries, authentication/authorization code paths, dependency manifests — and return a summary with risk hotspots and a recommended sweep split.
  2. Based on the recon summary, in ONE message issue parallel delegate calls to the sweeps it justifies: input-sweep (injection vectors, validation, output encoding, file uploads, redirects) and/or secrets-sweep (hardcoded credentials, secrets in logs and config, sensitive fields in API responses). Each instruction must be self-contained (the child shares none of your context): the exact files or directories to scan, what to flag with file:line locations, and severity hints.
  3. Do the deep exploit reasoning yourself, only on flagged locations: verify exploitability, build the proof of concept, and assign severity. A sweep's flag is a lead, not a finding — you own every verdict.
  4. Fold the verified findings into the Output Format below, marking which came from sweeps.

Read the full file on GitHub · 167 lines

Changes

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.

  1. 2d ago First seen · 167 lines · 35 tokens per session scan A 865690f84f48

Subscribe to this mod's changes

security-auditor is an agent published in the GitHub repository chankov/agent-fleet (10 stars, last pushed 6d ago), licensed MIT. It adds 35 tokens to every session and 1,768 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.