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/robinslange/learning-loop/seednpx skills add robinslange/learning-loop --skill seedgit clone --depth 1 https://github.com/robinslange/learning-loopWhat 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 | $0.00090 | $0.00886 |
| Opus 5 | $0.00045 | $0.00443 |
| Sonnet 5 | $0.00018 | $0.00177 |
| Haiku 4.5 | $0.00009 | $0.00089 |
Grade A, and why
seed 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 2d 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 — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Seed a fresh instance
Produces a seed-bundle-<date>/ an empty learning-loop instance can boot from via /learning-loop:init. Carries your working-style, not your knowledge or projects.
Paths
Resolve PLUGIN_DATA, VAULT, and the plugin root per ${CLAUDE_PLUGIN_ROOT}/skills-shared/paths-preamble.md (read it and apply). Resolve the auto-memory dir mechanically (do NOT hand-construct the slug):
node -e "import('${CLAUDE_PLUGIN_ROOT}/scripts/lib/memory-paths.mjs').then(m=>console.log(m.resolveMemoryDir(process.env.CLAUDE_PROJECT_DIR)))"
Process
1. Resolve inputs
- Default preset
--for-job:types = ["feedback"], no vault tiers. --types a,boverrides the type set.--tiers <tier>(e.g.3-permanent) opts in vault notes and triggers a hard scrub (see step 4).- Default name deny-list for the scrub: project-flavored feedback. Build it mechanically — do NOT hand-curate. Pass the patterns the operator confirms; start from obvious client/product/person tokens visible in filenames.
2. Mechanical selection
Run:
node ${CLAUDE_PLUGIN_ROOT}/scripts/seed-select.mjs <memDir> <types-csv> <deny-csv>
This returns {kept, dropped}. The type filter and name deny-list are mechanical. Do not add files the script dropped.
3. Present for consolidation (the home-brain benefit)
Show the operator the kept list and the dropped list with reasons. Ask: any kept file that is actually project-specific or personal and should be dropped? Any dropped file that is genuinely portable and should be kept? This is the consolidation moment — encourage pruning stale feedback at the source.
4. Scrub (only if --tiers used)
If vault tiers were opted in, run the candidate notes through the same scrubber harvest uses (pass PLUGIN_DATA so instance facts merge in):
node ${CLAUDE_PLUGIN_ROOT}/scripts/harvest-scrub.mjs "<denylistFile>" "<PLUGIN_DATA>" <note-path...>
Block anything the scrub blocks. (Reuses the harvest scrubber — same mechanical gate.) A vault tier can hold hundreds of notes; if the path list is large, pipe paths on stdin instead of argv: ... harvest-scrub.mjs "<denylistFile>" "<PLUGIN_DATA>" < notes.txt. For --for-job (no tiers) this step is skipped entirely.
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.
- 2d ago First seen · 53 lines · 90 tokens per session scan A 86098d6ef5cf
seed is a skill published in the GitHub repository robinslange/learning-loop (11 stars, last pushed 11d ago), licensed Apache-2.0. It adds 90 tokens to every session and 886 once invoked, about $0.0005 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…