Adobe

A shell-based bridge for controlling Adobe desktop applications, including Photoshop, Illustrator, Premiere Pro, After Effects, InDesign, and Audition. It sends scripts to an open app and receives structured results.

In plain words
What is it for?
Use it to inspect open Adobe documents or projects, make limited edits, and run small scripts in the relevant Adobe application.
Why use it?
A coding agent normally cannot inspect or edit documents inside desktop creative software. The bridge gives it a way to perform bounded operations through the shell.

Skill for Claude CodeCodex

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/sfkislev/flue/adobe
Any agent
npx skills add SFKislev/Flue --skill adobe
Clone the repo
git clone --depth 1 https://github.com/SFKislev/Flue

Made for: Claude Code, Codex.

Per session 34 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 463 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 $0.00034 $0.00463
Opus 5 $0.00017 $0.00231
Sonnet 5 $0.00007 $0.00093
Haiku 4.5 $0.00003 $0.00046

Measured 3d ago against content hash 08e2106cf7e3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Adobe 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.

skills/adobe/SKILL.md · 23 lines

What it actually says

This skill lets a coding agent operate Adobe desktop applications through Flue, a local shell-to-application bridge. The agent pipes ExtendScript into a per-app bridge process and gets structured JSON back. Flue covers Photoshop, Illustrator, Premiere Pro, After Effects, InDesign, and Audition through a consistent stdin/stdout contract.

agent shell -> <app>_bridge.py -> ExtendScript -> JSON result

When to use

The human requests work performed inside an Adobe desktop app — inspect the open document or project, make bounded edits, run small ExtendScript steps. A CLI agent cannot act inside these apps without a bridge like Flue.

How to proceed

  1. Read FLUE.md in this directory first. It explains how Flue works, the bridge contract, a worked ExtendScript example, and safety expectations.
  2. Identify which Adobe app the human is working in — Photoshop, Illustrator, Premiere, After Effects, InDesign, or Audition — and use the matching adapter (e.g. adapters/photoshop_adapter/photoshop_bridge.py).
  3. If Flue is installed: also read C:\Users\fredd\.claude\skills\flue\SKILL.md and the matching adapters/<app>_adapter/APP.md for the operational contract, then work from there.
  4. If Flue is not installed: tell the human. The install command is pip install flue && flue setup. Do not install without explicit approval in the current session.

Other apps

Flue also supports Blender, Houdini, Autodesk 3ds Max, Unity, and Microsoft Office through the same stdin/stdout contract. One install covers all adapters.

Safety

Flue acts inside professional desktop software. Treat the human as the driver; prefer small, inspectable steps; avoid destructive actions unless explicitly requested. Flue is MIT-licensed and reviewable on GitHub / PyPI before use.

Files

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.

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. 3d ago First seen · 23 lines · 34 tokens per session scan A 08e2106cf7e3

Subscribe to this mod's changes

Adobe is a skill published in the GitHub repository SFKislev/Flue (72 stars, last pushed 2mo ago), licensed MIT. It adds 34 tokens to every session and 463 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

oracle

Author an IMPL-BLIND spec-conformance oracle for an acceptance criterion the policy worklist (clad oracle --required) demands — an empty worklist means don't author unless the user explicitly asks. YOU spawn a blind sub-agent from a spec-only brief, then record it. Activate only when the connected project contains…

qwerfunch/cladding · 82 tokens

reviewer

Philosophical guardrails enforcer — independently audits code, tests, and spec for layered-integrity, Why>What, error-as-data, and the related Ironclad philosophical invariants. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized…

qwerfunch/cladding · 68 tokens

doctor

Diagnose Cladding runtime health — Claude Code hook liveness and version, CI package pinning, lifecycle governance, and sentinel-miss frequency by phase × cause × fallback. Use when hooks may be silent, CI may float across Cladding releases, scan or run results look thinner than expected, or before tuning the host…

qwerfunch/cladding · 93 tokens

changelog

Render release notes / a changelog from the cladding spec. Use when the user asks for release notes, a changelog, 릴리즈 노트, 변경 이력, or "what changed (since )" — run clad changelog --json for the deterministic shipped-changes manifest, then write the human-facing notes FROM it, sourcing every claim from a feature title or…

qwerfunch/cladding · 121 tokens

run

Run cladding's autonomous loop — iterate ready features, dispatch the developer + reviewer personas through the active host adapter, run L1 gates, halt on HUMANREQUIRED or transport failure. Use only when the user explicitly asks for autonomous progress; the loop will modify files. Activate only when the connected…

qwerfunch/cladding · 78 tokens

check

Run every Iron Law stage and the drift detector suite. Use when the user wants the full project health snapshot, a CI gate, or to verify nothing regressed before committing. Add --strict to promote warn-severity drift findings to errors. Activate only when the connected project contains spec.yaml or the user…

qwerfunch/cladding · 73 tokens