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 skills add 23blocks-OS/ai-maestro-plugins --skill agent-filesgit clone --depth 1 https://github.com/23blocks-OS/ai-maestro-pluginsWrote 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/23blocks-os/ai-maestro-plugins/agent-files)<a href="https://agentmods.dev/skills/23blocks-os/ai-maestro-plugins/agent-files"><img src="https://agentmods.dev/badge/skills/23blocks-os/ai-maestro-plugins/agent-files/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/23blocks-os/ai-maestro-plugins/agent-files"><img src="https://agentmods.dev/badge/skills/23blocks-os/ai-maestro-plugins/agent-files.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00100 | $0.01158 |
| Opus 5.5 | $0.00040 | $0.00463 |
| Sonnet 5.5 | $0.00020 | $0.00232 |
| Haiku 4.5 | $0.00010 | $0.00116 |
Grade A, and why
agent-files scanned grade A with 1 finding 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 15d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
compatibility: Requires curl 7.75 or newer, jq, openssl and shasum or sha256sum. Works on bash 3.2 (macOS) and Linux. A reachable S3-compatible store configured as a space. How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Files (AFP)
AFP gives agents one way to share files without a mounted folder. You put a file in a space (a bucket on an S3-compatible store), get back a reference, and send the reference. The receiver fetches the file from the same store and the SHA-256 is checked on the way. Every afp-*.sh command prints one JSON object and exits non-zero on failure, with {"ok":false,"error":{"code":...,"message":...}}.
Commands
afp-put.sh <file> [--space S] [--path P] [--ttl 7d] [--force] # store; prints ref, digest, size
afp-get.sh <ref | reference-object | @file> [--dest PATH] [--digest sha256:..] [--force]
afp-ls.sh [--space S] [--prefix P] [--limit N] # objects with owner, size, expiry, scan
afp-link.sh <ref> [--ttl 15m] # reference object with a time-limited url
afp-rm.sh <ref> [--admin] # delete object and manifest
afp-capabilities.sh [--space S] [--probe] # link, expire, versions, offline, ui
afp-config.sh add|list|remove|default ... # which spaces this machine can reach
A reference looks like afp://<space>/<path>. afp-put.sh without --space or --path uses the default space and <yyyy>/<mm>/<filename>. Downloads land in ~/.agent-files/downloads/<space>/<path> unless you pass --dest.
AFP or a message attachment
| The file | Use |
|---|---|
| Goes to one recipient with one message, under 25 MB | AMP attachment (amp-send.sh --attach) |
| Is large, shared by several parties, must outlive the message, or is for a person to pick up | AFP |
Sending and receiving a reference
Put the file, then send the reference object in the message so the receiver has the digest too. Until the AMP scripts send storage: afp attachments themselves, carry it in --context:
afp-put.sh report.pdf --space artifacts --ttl 7d # note ref and digest in the output
amp-send.sh alice "Report ready" "The Q3 report is in artifacts." --context '{"afp":{"ref":"afp://artifacts/2026/10/report.pdf","digest":"sha256:...","size":1827341}}'
What ships with it
1 file 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.
- 15d ago First seen · 59 lines · 100 tokens per session scan A 51b1d23fbf01
agent-files is a skill published in the GitHub repository 23blocks-OS/ai-maestro-plugins (11 stars, last pushed today), licensed MIT. It adds 100 tokens to every session and 1,158 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-05.
Other skills, from other repositories
memory-search
Runs read-only SQL over the local EvoFlux session database (evoflux.db) for structured analysis that ordinary recall cannot answer, such as aggregating tool failures across sessions, filtering turns by role, agent, or time range, reconstructing a complete execution chain, or checking a remembered claim against what a…
Read, create and manipulate PDF files — extract text and tables, merge, split, rotate, reorder and delete pages, read and fill AcroForm fields, add or strip metadata, encrypt and decrypt, and generate new PDFs from HTML or from scratch. Also covers rasterising pages to images so a PDF can actually be looked at, and…
intelligent-ui
Helps agents choose prose, native controls, dashboards, motion, or 3D from the user's goal. Use when a user asks to explain, teach, compare, or present information visually, asks you to choose a format, requests a video or code-authored animation as the deliverable, or an interactive artifact may improve understanding…
diagram-design
Creates polished, accessible SVG diagrams as EvoFlux Work widgets or project files. Use when a user asks to design, draw, redraw, or polish a software architecture, data flow, API sequence, state machine, ER model, deployment, service journey, or process diagram, or asks to turn Mermaid, draw.io, or Excalidraw content…
agents
Guidance for AI coding agents working on the adeptability (adept) codebase. Humans: see README.md for usage and CONTRIBUTING.md for the contributor workflow.
authoring-adept-agents
Write a good, portable adept agent (subagent): trigger-shaped description, one job per agent, generator/evaluator separation, explicit boundaries, restricted tools. Apply when creating or editing an agent file or running adept agent add.