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.
git clone --depth 1 https://github.com/snyk/saw-mcpWrote 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/rules/snyk/saw-mcp/saw_rules)<a href="https://agentmods.dev/rules/snyk/saw-mcp/saw_rules"><img src="https://agentmods.dev/badge/rules/snyk/saw-mcp/saw_rules/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/snyk/saw-mcp/saw_rules"><img src="https://agentmods.dev/badge/rules/snyk/saw-mcp/saw_rules.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.01602 | $0.01602 |
| Opus 5.5 | $0.00641 | $0.00641 |
| Sonnet 5.5 | $0.00320 | $0.00320 |
| Haiku 4.5 | $0.00160 | $0.00160 |
Grade A, and why
saw_rules 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 1mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Using curl or any HTTP client as a substitute How it starts
The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Snyk API & Web Project Rules
These rules define behavioral constraints and proactive behaviors for projects using SAW.
CRITICAL: Always Use the SAW MCP Server
For ANY task involving Snyk API & Web, SAW, or Probely, you MUST use the SAW MCP Server tools.
- Do NOT attempt to call the Snyk API & Web REST API directly via HTTP requests
- Do NOT write code that makes API calls to api.probely.com
- ALWAYS use the
probely_*MCP tools provided by the SAW MCP server - The SAW MCP server handles authentication, retries, and error handling for you
If a probely_* tool call fails, fix the parameters and retry
NEVER work around MCP tool failures by:
- Writing Python/shell scripts to call the Probely API directly
- Using curl or any HTTP client as a substitute
- Creating helper files to generate or fix payloads
The MCP tools handle formatting (e.g., pretty-printing login sequences). Bypassing them loses that.
Trigger Keywords
Use SAW MCP tools when the user mentions any of:
- "Snyk API & Web" / "Snyk APIWeb" / "Snyk API Web"
- "SAW" (when referring to security scanning)
- "Probely" (legacy name)
- "DAST" / "Dynamic Application Security Testing"
- "security scan" / "vulnerability scan" / "web scan"
- "target" (in security scanning context)
- "findings" / "vulnerabilities" (from scans)
Automatic Target Discovery
When a new project is loaded or workspace is opened, check if it's an application or API codebase (look for indicators like web frameworks, API routes, package.json with web dependencies, etc.).
Only for app/API projects:
- Search for corresponding SAW targets using
probely_list_targetswith the project name or domain - If a matching target is found, inform the user about existing security testing coverage
- If no target is found, suggest creating a new SAW target for the project
Skip this for non-app projects (libraries, CLI tools, documentation, etc.).
Destructive or Sensitive Tools
The following tools create, update, or delete resources. There is no server-side confirmation mechanism — the LLM must exercise caution and follow the safety rules in this document before calling them.
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.
- 1mo ago Changed · +2 tokens per session b8abbe4d7f23
- 2mo ago First seen · 159 lines · 1,600 tokens per session scan A d0fba6ac03fd
saw_rules is a cursor rule published in the GitHub repository snyk/saw-mcp (8 stars, last pushed 22d ago), licensed Apache-2.0. It adds 1,602 tokens to every session, about $0.0064 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other cursor rules, from other repositories
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.
control-plane-descriptors
Control plane descriptor and instance implementation patterns.
family-instance-domain-actions
Family instance domain action implementation patterns.