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 markmhendrickson/ateles --skill scrape-chatgpt-workoutgit clone --depth 1 https://github.com/markmhendrickson/atelesWrote 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/markmhendrickson/ateles/scrape-chatgpt-workout)<a href="https://agentmods.dev/skills/markmhendrickson/ateles/scrape-chatgpt-workout"><img src="https://agentmods.dev/badge/skills/markmhendrickson/ateles/scrape-chatgpt-workout.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00075 | $0.04016 |
| Opus 5 | $0.00037 | $0.02008 |
| Sonnet 5 | $0.00015 | $0.00803 |
| Haiku 4.5 | $0.00007 | $0.00402 |
Grade A, and why
scrape-chatgpt-workout 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.
How it starts
The opening of the file, as written. The whole thing — 377 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scrape ChatGPT Workout
Capture a ChatGPT Fitness GPT conversation stream, reconstruct workout sessions from both assistant summaries and raw user messages, then store as workout_session entities in Neotoma.
When to Use
- User says "scrape chatgpt workout", "import chatgpt fitness", "backfill workouts from chatgpt"
- User provides a ChatGPT conversation URL (chatgpt.com/…/c/…)
- User wants to backfill historical workout sessions from a ChatGPT fitness log
Prerequisites
- Claude in Chrome extension connected (required for fetch interception)
- ChatGPT tab open and logged in
- Neotoma MCP available
Overview
The skill has two phases:
- Capture — Install a fetch interceptor in the ChatGPT tab, trigger conversation API re-fetch, buffer the full streaming response (~4MB for a long conversation)
- Reconstruct & Store — Parse the captured mapping tree, reconstruct sessions from assistant summaries + raw user messages, store each session to Neotoma
Phase 0: Cache Check
Before touching the browser, check whether a conversation entity for this chat already exists in Neotoma:
retrieve_entities(entity_type="conversation", search="chatgpt-fitness-gpt-254868b9")
Or by URL pattern:
retrieve_entity_by_identifier(identifier="254868b9-73d0-8329-b395-c48b6b8a8fef", entity_type="conversation")
If a matching entity exists and a linked file_asset (the JSONL transcript) is present, retrieve the file URL and load the messages from it instead of re-scraping. This allows re-analysis without re-opening the ChatGPT tab.
retrieve_file_url(entity_id="<file_asset_entity_id>")
# → download JSONL and parse into _allMessages equivalent
Only proceed to Phase 1 if no cached transcript is found, or if the user explicitly requests a fresh capture.
Phase 1: Capture
Step 1.1 — Confirm tab
Use tabs_context_mcp to confirm the ChatGPT conversation tab is open. If not, ask the user to navigate to the conversation URL first.
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 · 377 lines · 75 tokens per session scan A 279c4dcb8305
scrape-chatgpt-workout is a skill published in the GitHub repository markmhendrickson/ateles (6 stars, last pushed today), licensed MIT. It adds 75 tokens to every session and 4,016 once invoked, about $0.0004 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 2026-09-03.
Other skills, from other repositories
browser-use
Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work.
life-automation-audit
Use this as an on-demand long prompt. Do not paste it into SOUL.md / PRINCIPLES.md / AGENTS.md. When the user asks to audit their life for automation opportunities, load and follow this prompt.
browser-automation-studio
Reference for using Browser Automation Studio (BAS) to execute browser workflows, validate UI behavior, and debug automation failures.
demo-walkthrough
Use when the user needs a product walkthrough from authorized browser automation, with Axon used for public page scrape/screenshot evidence or captured URLs.
knowledge-ingest
Use when indexing docs portals where Axon Chrome rendering, capture-time automation scripts, or host browser-assisted URL discovery are needed.
discord
Control Discord via the discord tool: manage messages, reactions, polls, channels, and bot presence.