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/luckypipewrench/pipelock/agents-mdgit clone --depth 1 https://github.com/luckyPipewrench/pipelockWrote 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/instructions/luckypipewrench/pipelock/agents-md)<a href="https://agentmods.dev/instructions/luckypipewrench/pipelock/agents-md"><img src="https://agentmods.dev/badge/instructions/luckypipewrench/pipelock/agents-md.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 | $0.02745 | $0.02745 |
| Opus 5 | $0.01373 | $0.01373 |
| Sonnet 5 | $0.00549 | $0.00549 |
| Haiku 4.5 | $0.00275 | $0.00275 |
Grade A, and why
pipelock 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 today.
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 — 208 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md - Pipelock Contributor Guide
Pipelock is an agent firewall: a network and tool proxy that mediates AI-agent HTTP, WebSocket, and MCP traffic and scans it for secret exfiltration, prompt injection, SSRF, and tool poisoning. Direct egress controls are deployment guidance; binary-enforced coverage applies to mediated traffic.
Quick Reference
| Item | Value |
|---|---|
| Module | github.com/luckyPipewrench/pipelock |
| Go | 1.25+; CI tests Go 1.25 and 1.26 |
| License | Apache 2.0 core, ELv2 under enterprise/ |
| Binary | Single Go binary; size varies by OS, build tags, and release flags |
| Dependencies | See go.mod. Run make stats before citing the current direct-dependency count. |
Capability surface
The code-checked capability manifest is docs/security/capability-manifest.json. It lists the operator entry point and license gate for each surface below. Free means no license feature is required.
| Capability | Access | Operator entry point | Platform or deployment qualifier |
|---|---|---|---|
| HTTP, WebSocket, and proxy mediation | Free | pipelock run | Applies only to traffic routed through Pipelock; direct agent egress needs containment or network policy. |
| MCP proxy scanning | Free | pipelock mcp proxy | Applies only to MCP transports that run through the Pipelock proxy. |
| Signed action receipts | Free | flight_recorder | |
| Host containment for agent processes | Free | pipelock contain | Kernel-enforced containment requires Linux, nftables, and the managed identities; other targets cannot provide this containment path. |
| Single-agent process sandbox | Free | pipelock sandbox | Kernel isolation requires Linux user namespaces. Without them, best-effort network isolation only sets HTTP(S)_PROXY; seccomp adds restrictions only on linux/amd64. |
| Global canary tokens | Free | canary_tokens | |
| Named agent profiles | Pro | agents. | |
| Per-agent sandbox overrides | Pro | agents..sandbox | |
| Per-agent crypto address allowlists | Pro | agents..allowed_addresses | |
| Read-only operator dashboard | Pro or Enterprise | pipelock dashboard serve | |
| Full assessment artifacts | Assess | pipelock assess finalize | |
| Conductor fleet coordination | Enterprise | pipelock conductor serve | |
| Per-agent coverage certificates | Pro | pipelock dashboard coverage-cert generate | |
| Exemption lifecycle records | Pro | pipelock dashboard exemption list | |
| Legal-hold metadata | Pro | pipelock dashboard legal-hold list | |
| Fleet audit sink | Enterprise | pipelock fleet-sink | |
| Fleet receipt reports | Enterprise | pipelock conductor fleet report | |
| Emergency kill switch | Free | kill_switch | |
| Rule bundles | Free | rules | |
| Adaptive enforcement | Free | adaptive_enforcement | |
| TLS interception | Free | tls_interception | Requires a configured local CA that the intercepted client trusts; passthrough traffic remains encrypted and cannot be body-scanned. |
| A2A protocol scanning | Free | a2a_scanning | Applies to A2A traffic carried on Pipelock's supported forward-proxy and MCP HTTP paths. |
| Browser shield | Free | browser_shield | Applies only to supported browser response traffic that flows through Pipelock. |
| Offline receipt verification | Free | pipelock verify-receipt | |
| Verdict explanations | Free | pipelock explain | |
| Deployment diagnostics | Free | pipelock doctor |
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.
- today Changed · +5 lines · +76 tokens per session bc14bed9ac4a
- 4d ago First seen · 203 lines · 2,669 tokens per session scan A 0118d282ee3f
pipelock AGENTS.md is an instructions file published in the GitHub repository luckyPipewrench/pipelock (832 stars, last pushed today), licensed Apache-2.0. It adds 2,745 tokens to every session, about $0.0137 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
arcjet-js AGENTS.md
Instructions for arcjet/arcjet-js, covering agent guidance, examples live in arcjet/examples and integration work: review before a pr.
helm-ai-kernel AGENTS.md
AGENTS.md instructions for Mindburn-Labs/helm-ai-kernel, covering agent operational guidelines for helm-ai-kernel, developer runbook and governance & rules.
Agent-Security-Regression-Harness AGENTS.md
Instructions for OWASP/Agent-Security-Regression-Harness, covering ai agent guide: owasp agent security regression harness, project overview, architecture, contribution workflow and what needs doing.
ai-appsec AGENTS.md
Instructions for subodhkc/ai-appsec, covering agents.md — haiec agent security mcp (phase -1 → phase 18), 1. workspace layout, 2. architectural principles (apply to every phase), 3. hard constraints (never violate) and 4. phase discipline.
enforra AGENTS.md
Instructions for enforra/enforra, covering agents.md, enforra engineering doctrine, package boundaries, required workflow before pr and diagnostics.
promptsonar AGENTS.md
Instructions for meghal86/promptsonar, covering promptsonar agent instructions, operating principles, repo layout, common commands and change workflow.