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/lamenting-hawthorn/skillloop/goal-loopnpx skills add lamenting-hawthorn/SkillLoop --skill goal-loopgit clone --depth 1 https://github.com/lamenting-hawthorn/SkillLoopWhat 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 | $0.00049 | $0.03191 |
| Opus 5 | $0.00024 | $0.01596 |
| Sonnet 5 | $0.00010 | $0.00638 |
| Haiku 4.5 | $0.00005 | $0.00319 |
Grade A, and why
goal-loop 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 3d 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 — 354 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Goal Loop — The /goal Pattern
The /goal primitive: a cron job that runs on a cadence until an objective
condition is met, verified by an independent checker model — not the agent
that writes the code. This is the maker-vs-checker split applied to the stop
condition itself.
When to Use
- After passing the
pre-loop-checklist(4-condition test + 30-second check) - When you have a clear, machine-checkable goal condition
- When the work is iterative (agent tries, fails, learns, retries)
- When verification can be fully automated
Do NOT use for:
- One-shot tasks (use a single
delegate_task) - Vague goals ("make it better")
- Work that can't be verified by a test/build/linter
- Anything where "done" is a judgment call
The Pattern (4 Parts)
┌──────────────────────┐
│ Cron job fires on │
│ cadence (e.g. 30m) │
└──────────┬───────────┘
│
┌──────────▼───────────┐
│ Read STATE.md │
│ (resume, don't │
│ restart from zero) │
└──────────┬───────────┘
│
┌──────────▼───────────┐
│ CHECKER subagent │
│ (DIFFERENT model) │
│ Evaluate: is goal │
│ condition met? │
└──────────┬───────────┘
│
┌───────────┴───────────┐
│ │
▼ ▼
┌──────────┐ ┌──────────┐
│ GOAL │ │ GOAL │
│ MET ✓ │ │ NOT MET │
└────┬─────┘ └────┬─────┘
│ │
▼ ▼
┌──────────┐ ┌──────────┐
│ Report │ │ WORKER │
│ success │ │ subagent │
│ Update │ │ (does │
│ STATE │ │ the │
│ Option: │ │ work) │
│ self- │ └────┬─────┘
│ cancel │ │
└──────────┘ ▼
┌──────────┐
│ Update │
│ STATE │
│ Report │
│ progress│
│ (loop │
│ again │
│ next │
│ tick) │
└──────────┘
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.
- 3d ago First seen · 354 lines · 49 tokens per session scan A 0a949de1ee12
goal-loop is a skill published in the GitHub repository lamenting-hawthorn/SkillLoop (9 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 49 tokens to every session and 3,191 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.
Other skills, from other repositories
fine-tuning-with-trl
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
mnemosyne-memory-override
Hard rule override that forces Mnemosyne for all durable memory storage. The legacy memory tool is DEPRECATED for user preferences, credentials, and project conventions. Use memory ONLY for ephemeral session state.
openlore
Query and publish to an OpenLore knowledge base over SSH using ordinary shell commands. Use when a task needs project documentation, runbooks, shared team knowledge, or a place to publish findings.
github-code-review
Review PRs: diffs, inline comments via gh or REST.
plan
Plan mode: write markdown plan, no execution.
simplify-code
Sequential 3-lens cleanup of recent code changes.