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/ciroaurelio22/agentloop/updatenpx skills add ciroaurelio22/AgentLoop --skill updategit clone --depth 1 https://github.com/ciroaurelio22/AgentLoopWhat 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 | $0.00033 | $0.00804 |
| Opus 5 | $0.00016 | $0.00402 |
| Sonnet 5 | $0.00007 | $0.00161 |
| Haiku 4.5 | $0.00003 | $0.00080 |
Grade A, and why
agent-loop-update scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Loop — Update
Pull the latest kit from GitHub (https://github.com/ciroaurelio22/AgentLoop.git) and reinstall into this repository, preserving queue and task programs.
Run the updater
From the target repo root:
pnpm agent:update
Or directly:
node scripts/agent/update.mjs --target .
The script:
- Clones or
git pulls the kit into%TEMP%/agent-loop-kit(or$TMPDIR/agent-loop-kit) - Runs
bin/install.mjs --target . --all --force --updatedfrom that fresh copy
Optional flags:
node scripts/agent/update.mjs --target . --branch master
node scripts/agent/update.mjs --target . --kit-dir "C:/path/to/cache"
Restart Agent Console (required)
After every update, restart the web UI. The kit replaces files on disk (tools/agent-gui/, scripts, hooks), but a server already listening on port 9477 keeps the old code in memory. pnpm agent:gui:ensure alone does not restart a healthy running instance.
- Stop the current Agent Console:
- Close the terminal that runs
pnpm agent:gui, or - Kill the PID stored in
.agent-loop/gui.pid(Windows:taskkill /F /PID <pid>, Unix:kill <pid>)
- Close the terminal that runs
- Start it again:
pnpm agent:gui:ensure
Or open it manually:
pnpm agent:gui
- Reload the browser tab at
http://127.0.0.1:9477(hard refresh if needed).
Always tell the user to restart Agent Console after an update completes.
Preserved user data
The installer does not overwrite:
.agent-loop/queue.json.agent-loop/scratchpad.md.agent-loop/autostartspecs/agent-tasks/TASK-*.md
Verify after update
pnpm agent:status
# restart Agent Console (see above), then:
pnpm agent:gui:ensure
Report what changed (scripts, hooks, GUI, skills), confirm queue/tasks are intact, and remind the user to reload the browser after the GUI restart.
Constraints
- Source of truth is the GitHub repo — do not hand-patch kit files in the target repo.
- Do not overwrite user task programs or queue state.
- Requires git in PATH for clone/pull.
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 · 106 lines · 33 tokens per session scan A 0d80c4abfa05
agent-loop-update is a skill published in the GitHub repository ciroaurelio22/AgentLoop (10 stars, last pushed 2mo ago), licensed MIT. It adds 33 tokens to every session and 804 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
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…