cleanup-transcript

A command for improving video or audio transcripts in SRT subtitle or plain-text format. It examines the structure, fixes likely transcription mistakes, and asks questions when the meaning is unclear.

In plain words
What is it for?
Use it to clean subtitles, correct names and technical terms, identify speakers and topics, and review uncertain parts of a transcript.
Why use it?
It reduces the manual work of turning a rough transcript into readable, accurate text while preserving the conversation’s structure.

Command

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/technicalpickles/pickled-claude-plugins/cleanup-transcript
Clone the repo
git clone --depth 1 https://github.com/technicalpickles/pickled-claude-plugins
Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 816 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 $0.00026 $0.00816
Opus 5 $0.00013 $0.00408
Sonnet 5 $0.00005 $0.00163
Haiku 4.5 $0.00003 $0.00082

Measured yesterday against content hash 8ab98de05c3f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cleanup-transcript 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 yesterday.

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.

plugins/dev-tools/commands/cleanup-transcript.md · 121 lines

How it starts

The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Transcript Cleanup

Clean up the transcript for readability and accuracy. Supports both SRT subtitle files and plaintext transcripts.

Step 1: Read and Detect Format

Read the transcript file provided as $1.

Detect format:

  • SRT format: Has numbered entries with timestamps like 00:01:23,456 --> 00:01:25,789
  • Plaintext: Raw text without timestamps

For SRT files, strip the entry numbers, timestamps, and blank lines to extract the raw text. Note approximate time ranges for major topic shifts if helpful for the user.

Step 2: Identify High-Level Structure

Analyze the content and identify:

  1. Major topics/sections - What distinct subjects are discussed?
  2. Approximate flow - How does the conversation progress?
  3. Number of speakers - Is this a monologue or conversation?

Present this structure to the user for confirmation before proceeding.

Step 3: Identify Transcription Errors (Iterative)

Scan for likely transcription errors. Common patterns include:

Proper Nouns

  • Tool names (often mangled: "get hub" → GitHub, "chat GBT" → ChatGPT)
  • Company/product names
  • People's names
  • Technical terms and acronyms

Commands and Code

  • CLI commands (often split or garbled)
  • File paths
  • Code snippets mentioned verbally

Domain Terminology

  • Industry-specific terms
  • Project-specific vocabulary
  • Abbreviations

For each category of errors found:

  1. Present the suspected errors with surrounding context
  2. Ask the user to confirm corrections
  3. Note any additional context they provide
  4. Proceed to the next category

If context files were provided ($2, $3, etc.), read them to help resolve ambiguities. Context files might include:

  • Session logs (JSONL files from Claude sessions)
  • Related documentation
  • Glossaries or term lists

Step 4: Clarify Speaker Attribution

If multiple speakers are detected:

  1. Ask how many speakers there are
  2. Ask for their names/identifiers
  3. Present ambiguous exchanges and ask for attribution

Read the full file on GitHub · 121 lines

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. yesterday First seen · 121 lines · 26 tokens per session scan A 8ab98de05c3f

Subscribe to this mod's changes

cleanup-transcript is a command published in the GitHub repository technicalpickles/pickled-claude-plugins (10 stars, last pushed 4d ago), licensed MIT. It adds 26 tokens to every session and 816 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.