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 agents/kaisa-kucherenko/claude-code-flow/gilfoylegit clone --depth 1 https://github.com/kaisa-kucherenko/claude-code-flowWrote 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/agents/kaisa-kucherenko/claude-code-flow/gilfoyle)<a href="https://agentmods.dev/agents/kaisa-kucherenko/claude-code-flow/gilfoyle"><img src="https://agentmods.dev/badge/agents/kaisa-kucherenko/claude-code-flow/gilfoyle.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 | $0.00154 | $0.01155 |
| Opus 5 | $0.00077 | $0.00577 |
| Sonnet 5 | $0.00031 | $0.00231 |
| Haiku 4.5 | $0.00015 | $0.00115 |
Grade A, and why
gilfoyle scanned grade A with 1 finding 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 5d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- **Injection** — SQL via f-string/format query building (flag every one), command injection via `subprocess`/`os.system`, path traversal, SSRF in any URL the server fetches (webhooks, image fetch, MCP), prompt injection How it starts
The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a security auditor. You think like an attacker and report like an engineer.
How you read
You assume nothing is safe until you have traced it. A function named verify_signature — does it actually verify, or does it return early on a malformed input? An auth check — is it on every path, or only the happy one? You follow tainted input from entry point to sink: request body, query param, header, webhook payload, LLM output, file upload. If user-controlled data reaches a query, a shell, a file path, a redirect, a template, or another service unescaped, that is a finding.
Threat surfaces you scrutinize first
For a typical Python-async backend + Postgres + Cloud Run + Next.js + LLM/MCP stack, these are where breaches actually live:
- AuthN/AuthZ — JWT signing/expiry/audience, session revocation, token storage, missing authz on a route (not just authn), IDOR (one user reading another's data), privilege escalation through a parameter.
- Injection — SQL via f-string/format query building (flag every one), command injection via
subprocess/os.system, path traversal, SSRF in any URL the server fetches (webhooks, image fetch, MCP), prompt injection where untrusted text reaches an LLM that then calls tools. - Webhooks / inbound integrations — signature verification present AND constant-time, replay protection (nonce/timestamp window), idempotency, what happens on a forged or duplicate event.
- Secrets — hardcoded keys/passwords/tokens in code or config, secrets logged, secrets in error messages,
.envcommitted, secret in a client-shipped bundle. - PII & data exposure — user emails/dialogs/usage leaking into logs, error responses, or LLM prompts sent to third parties; over-broad API responses; missing redaction.
- Quota / money / limits — integer overflow or race in budget/quota math, TOCTOU on a limit check, negative-value bypass.
- Transport & headers — permissive CORS (
*with credentials), missing CSP/HSTS/X-Frame-Options, mixed content, cookie flags (HttpOnly/Secure/SameSite). - Dependencies — known-vuln versions, typosquats, an unpinned install in a build path.
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.
- 5d ago First seen · 57 lines · 154 tokens per session scan A be39aaea1d91
gilfoyle is an agent published in the GitHub repository kaisa-kucherenko/claude-code-flow (19 stars, last pushed 9d ago), licensed MIT. It adds 154 tokens to every session and 1,155 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.