agent-tools

A guide for debugging and extending the tools used by Open Edit, an agent system for creating and editing media. It covers the sandbox, reusable skills, saved style preferences, asset search, and tool routing.

In plain words
What is it for?
Use it when changing agent tools, sandbox execution, style memory, asset search, rendering, or the central tool dispatcher.
Why use it?
It helps developers find the right part of the system without breaking safety controls or creating assets that should have been searched for first.

Skill for Claude CodeCodexCursor

Install

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.

agentmods
npx agentmods add skills/ah64-dll/openedit/agent-tools
Any agent
npx skills add AH64-dll/OpenEdit --skill agent-tools
Clone the repo
git clone --depth 1 https://github.com/AH64-dll/OpenEdit

Made for: Claude Code, Codex, Cursor.

Per session 49 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 490 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00049 $0.00490
Opus 5 $0.00024 $0.00245
Sonnet 5 $0.00010 $0.00098
Haiku 4.5 $0.00005 $0.00049

Measured 2d ago against content hash 4b7bb71a20fa, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-tools 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.

.cursor/skills/agent-tools/SKILL.md · 40 lines

What it actually says

Open Edit — Agent Tools

Orient

  1. Read architecture/BLUEPRINT.md (Agent Tools band).
  2. Prefer host playbooks under skills/ over exploring open_edit/**.
  3. Runtime libraries live in open_edit/agent/skills/; LLM-facing Markdown skills live in skills/.

Rules

  • Keep the sandbox — required for run_python / run_script / free-form and motion-graphics run_render. Improve diagnostics; do not remove.
  • Search before generate — call search_assets (+ import_asset) for stock video/photo/audio before generate_visual_for_segment / music / SFX generate when licensed stock fits.
  • Style memory — follow skills/style-memory.md: read profile early; capture confirmed user preferences via capture_style_hint / pins; do not silently invent style.
  • One dispatcher — serve / MCP / Pi all go through kernel.tool_executoragent.tools.TOOL_TABLE.

Symptom → package

Symptom Go to
run_script / free-form fails agent/sandbox/ (bridge, backends, bootstrap, staging)
Silence / narrative / music / SFX wrong agent/skills/ + matching pyagent_*
Style forgotten across turns style/, agent/style_inject.py, skills/style-memory.md
Tool call errors kernel/tool_executor.pyagent/tools/pyagent_*.py
Stock search bad agent/tools/pyagent_search_assets.py

Do not

  • Explore source to rediscover pillar tools — use skills/open-edit-mcp.md / tool_surface.md.
  • Strip sandbox “for safety later.”
  • Edit open_edit/harness_skills/ directly — edit skills/ then sync copies.
Changes

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.

  1. 2d ago First seen · 40 lines · 49 tokens per session scan A 4b7bb71a20fa

Subscribe to this mod's changes

agent-tools is a skill published in the GitHub repository AH64-dll/OpenEdit (46 stars, last pushed 12d ago), licensed MIT. It adds 49 tokens to every session and 490 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 2026-08-30.

Related

Other skills, from other repositories

video-to-llm

Read a local video file — lecture, course, recording, screencast — by turning it into one timestamped document, once, and reusing it for every question afterwards. Use when the user points at a video file or folder of videos and wants its contents understood, searched, summarised, quoted, or checked. Also use to…

navdeep-h-singh/video-to-llm · 107 tokens

video-prep

Runs the tedious cleanup pass on raw video recordings before the real edit — cut silent gaps, transcribe subtitles in any language (Mandarin by default), normalize speed, and stitch clips in natural numeric order. Use when the user drops recordings into a raw/ folder or asks to "clean up", "prep", or "edit" their…

lucyfengcl-art/video-prep · 124 tokens

hr-onboarding

A new-hire onboarding plan as a single page — first week schedule, buddy + manager intro, learning track, equipment checklist, and "you're set when…" outcomes. Use when the brief mentions "onboarding", "new hire", "first week plan", or "入职".

nexu-io/open-design · 62 tokens

pr-to-video

Turn a GitHub pull request (a PR URL, owner/repo#N, or 'this PR' in a checked-out repo) into a code-change explainer video — changelog, feature reveal, fix, or refactor walkthrough built from the diff, commits, and files: the input is a code change, not a website. Not a product promo (/product-launch-video) or a no-PR…

heygen-com/hyperframes · 104 tokens

general-video

Author or edit a custom HyperFrames composition when no specialized workflow fits, or when BRIEF.md sets flow: companion. Use for longer or multi-scene pieces, brand and sizzle reels, montages, static loops, static title cards, footage remixes, and freeform builds. Use motion-graphics instead for a short unnarrated…

heygen-com/hyperframes · 92 tokens

motion-graphics

A short, design-led motion graphic where motion is the message — kinetic typography, stat count-up, chart/data-viz hit, logo sting / brand lockup, lower-third / callout / social overlay, animated map (highlight regions, connect places, zoom to a location), animated tweet / news-article / headline, webpage / UI…

heygen-com/hyperframes · 139 tokens