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 skills/adrianleb/hammock-skill/hammocknpx skills add adrianleb/hammock-skill --skill hammockgit clone --depth 1 https://github.com/adrianleb/hammock-skillWrote 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/skills/adrianleb/hammock-skill/hammock)<a href="https://agentmods.dev/skills/adrianleb/hammock-skill/hammock"><img src="https://agentmods.dev/badge/skills/adrianleb/hammock-skill/hammock.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 5.1 | $0.00087 | $0.01740 |
| Opus 5 | $0.00044 | $0.00870 |
| Sonnet 5 | $0.00017 | $0.00348 |
| Haiku 4.5 | $0.00009 | $0.00174 |
Grade A, and why
Hammock 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 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.
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 — 261 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hammock Driven Development
A methodology for solving hard problems through deliberate thinking. Most bugs come from misconception, not typos. The cheapest place to fix bugs is during design.
Interactive Process
Use AskUserQuestion tool throughout this process to gather information efficiently. Batch related questions together (max 4 per call). This creates a structured dialogue rather than free-form back-and-forth.
Quick Start
Use AskUserQuestion to determine mode and project type:
Question 1: "What type of design work is this?"
Header: "Project type"
Options:
- New project: "Starting something from scratch"
- New feature: "Adding new capability to existing system"
- Bug investigation: "Finding root cause of an issue"
- Architecture decision: "Significant structural change"
- Refactor: "Improving existing code structure"
- Plan review: "Review an existing design or proposal"
Question 2: "How deep should we go?"
Header: "Mode"
Options:
- Quick: "Problem → Understand → Tradeoffs (time pressure)"
- Standard (Recommended): "Full process without hammock pauses"
- Deep: "Include hammock time prompts for major decisions"
The Process
Phase 1: State the Problem
Use AskUserQuestion:
Question: "What problem are we actually solving? (Describe the pain, not the feature)"
Header: "Problem"
Options:
- [Let user type - use "Other" flow]
Question: "How will we know this is solved?"
Header: "Success"
Options:
- [Let user type - use "Other" flow]
If user gives a feature instead of a problem, probe deeper:
Question: "You mentioned [feature]. What pain or issue does this address?"
Header: "Root problem"
Write the problem statement explicitly before proceeding.
Phase 2: Understand the Problem
First, explore the codebase silently using Glob, Grep, Read tools. Then use AskUserQuestion to fill gaps:
Question: "What constraints must we respect?"
Header: "Constraints"
Options:
- Performance critical: "Has latency/throughput requirements"
- Security sensitive: "Handles auth, PII, or secrets"
- Backward compatible: "Must not break existing clients"
- Time constrained: "Has a deadline"
multiSelect: true
Question: "What's unclear that we should research?"
Header: "Unknowns"
Options:
- [Based on exploration findings, offer specific unknowns discovered]
- [Or let user type]
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 261 lines · 87 tokens per session scan A 5da89b634e50
Hammock is a skill published in the GitHub repository adrianleb/hammock-skill (11 stars, last pushed 6mo ago), licensed MIT. It adds 87 tokens to every session and 1,740 once invoked, about $0.0004 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 skills, from other repositories
reviewing-unit-tests
Use in a dedicated subagent for post-TDD blind unit-test review against requirements and public contracts to find missing cases, wrong assertions, and brittle tests.
behavior-driven-development
Use when implementing any feature or bugfix, before writing implementation code. Triggered by keywords: TDD, BDD, test-first, test-driven, write-test-first.
requesting-code-review
Use in a dedicated subagent when completing tasks, implementing major features, or doing post-TDD code review before merge to verify work meets requirements.
skill-creator
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
resilient-coding-agent
Run long-running coding agents (Codex, Claude Code, etc.) in tmux sessions that survive orchestrator restarts, with automatic resume on interruption.
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.