What is ghost-decode?

A command that extracts text hidden in moving dots or noise in a video. It uses a bundled decoder and checks the resulting image for the recovered words.

Use it to decode an MP4, MOV, AVI, or WEBM file and save the recovered text, a clean image, and a heatmap showing faint details.

It removes the need to manually inspect noisy video frames or set up a separate decoding process.

ghost-decode — Command for coding agents by haroontrailblazer: A command that extracts text hidden in moving dots…

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 commands/haroontrailblazer/ghost-font-decoder/ghost-decode

Clone the repo

git clone --depth 1 https://github.com/haroontrailblazer/ghost-font-decoder

Or install ghost-font-decoder, the plugin that ships this one along with the rest of its 1 skill, 1 command.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/haroontrailblazer/ghost-font-decoder/ghost-decode.svg)](https://agentmods.dev/commands/haroontrailblazer/ghost-font-decoder/ghost-decode)

Your own site

<a href="https://agentmods.dev/commands/haroontrailblazer/ghost-font-decoder/ghost-decode"><img src="https://agentmods.dev/badge/commands/haroontrailblazer/ghost-font-decoder/ghost-decode.svg" alt="Measured on agentmods" height="20"></a>

Per session 10 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 386 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.

Model	Per session	Once invoked
Fable 5	$0.00010	$0.00386
Opus 5	$0.00005	$0.00193
Sonnet 5	$0.00002	$0.00077
Haiku 4.5	$0.00001	$0.00039

Measured 4d ago against content hash 7540cf8e586a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ghost-decode 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 4d 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.

commands/ghost-decode.md · 40 lines

What it actually says

Immediately decode a ghost-font video (text hidden in moving dots / noise) by running the bundled decoder. Do the work now — do NOT ask the user to set up a folder, install the plugin differently, or point at files beyond the video path.

Target video: $ARGUMENTS

If $ARGUMENTS is empty, use the most recently modified .mp4, .mov, .avi, or .webm file in the current directory. Only ask the user if there is more than one plausible candidate.

Then, without further prompting:

Check deps: python -c "import cv2, numpy". If it fails, run pip install -r "${CLAUDE_PLUGIN_ROOT}/requirements.txt".
Run the decoder (it writes revealed.png and revealed_heatmap.png):
```
python "${CLAUDE_PLUGIN_ROOT}/decode.py" "<video-path>" -o "<scratch-out-dir>"
```
${CLAUDE_PLUGIN_ROOT} is the plugin's install directory. On Windows PowerShell it is $env:CLAUDE_PLUGIN_ROOT; if it expands empty, substitute the install path literally.
Read revealed.png with vision to confirm the text (the heatmap revealed_heatmap.png shows faint detail the clean mask may drop).
Reply with the recovered text on its own line, plus the output image paths:
```
Text in the video: **<DECODED TEXT>**
```

Never OCR a raw frame — every frame is uniform noise; the message is only in the motion. Do not claim success unless the decoder actually ran and you inspected revealed.png.

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.

4d ago First seen · 40 lines · 10 tokens per session scan A 7540cf8e586a

Subscribe to this mod's changes

ghost-decode is a command published in the GitHub repository haroontrailblazer/ghost-font-decoder (10 stars, last pushed 1mo ago), licensed MIT. It adds 10 tokens to every session and 386 once invoked, about $0.0001 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.

Research claude-ai claude-code-plugin claude-skills codex-plugin plugins

ghost-decode

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

What it actually says

What this file has done since we first saw it

Other commands, from other repositories

post

show

audit-project-github

sdlc

claudikins-kernel:verify

send