Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/oborchers/fractional-ctonpx agentmods add commands/oborchers/fractional-cto/compress-allWrote 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/commands/oborchers/fractional-cto/compress-all)<a href="https://agentmods.dev/commands/oborchers/fractional-cto/compress-all"><img src="https://agentmods.dev/badge/commands/oborchers/fractional-cto/compress-all/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/oborchers/fractional-cto/compress-all"><img src="https://agentmods.dev/badge/commands/oborchers/fractional-cto/compress-all.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00021 | $0.01030 |
| Opus 5.5 | $0.00008 | $0.00412 |
| Sonnet 5.5 | $0.00004 | $0.00206 |
| Haiku 4.5 | $0.00002 | $0.00103 |
Grade A, and why
compress-all 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 1mo 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 — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Recursively find and compress all files matching the given filename using the markdown-compression skill. Always runs in auto-approve mode with batched section processing.
Follow this process exactly:
Step 1: Discover Files
Use the Glob tool to search for **/<filename> in the current working directory, where <filename> is the argument from $ARGUMENTS (e.g., CLAUDE.md).
If no filename was provided, use AskUserQuestion to ask which filename to search for (e.g., CLAUDE.md, ARCHITECTURE.md, TREE.md).
If no files are found, inform the user and stop.
Step 2: Confirm File List
For each discovered file, read it and count approximate tokens (words * 1.3).
Present the file list:
| # | File | Tokens | Sections |
|---|------|--------|----------|
| 1 | ./CLAUDE.md | ~850 | 12 |
| 2 | ./src/CLAUDE.md | ~320 | 5 |
| 3 | ./supabase/CLAUDE.md | ~1,200 | 18 |
Total: ~X tokens across N files.
Use AskUserQuestion to confirm:
- Compress all — process every file in the list
- Exclude some — let the user specify files to skip (then confirm the reduced list)
Step 3: Determine Mode
If --lossless was passed in arguments, use lossless mode for all files. Otherwise, default to lossy mode (no question needed — lossy is the recommended default for batch operations).
Step 4: Process Each File
For each file in the confirmed list, in order:
4a: Announce
Show a header for the current file:
## Compressing [N]/[total]: [file-path]
4b: Pre-Analysis
Read the file, parse heading hierarchy, split into sections, flag structural issues. Show a brief section table (same format as /markdown-compressor:compress Step 3).
4c: Batched Compression
Process sections using the same batched approach as /markdown-compressor:compress Step 4, with auto-approve always active:
- For each batch of up to 5 non-empty sections:
- Dispatch all
section-compressoragents in the batch in a single message (parallel Agent tool calls) - Agents are read-only (Read/Grep/Glob only) and receive section text in their prompt — they do not read or modify the target file. It is safe to dispatch all agents in a batch simultaneously.
- Wait for all to return
- Dispatch all
- If in lossy mode, dispatch all
compression-revieweragents in the batch in a single message (parallel Agent tool calls). Wait for all to return. Incorporate critical fixes automatically. - Show batch summary (one line per section):
Section [N]/[total]: [heading] — ~X → ~Y tokens (-Z%) - Write all compressed sections in the batch to the file sequentially (top to bottom) using the Edit tool.
- Re-read the file before processing the next batch.
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.
- 1mo ago First seen · 111 lines · 21 tokens per session scan A dbe3c86134a5
compress-all is a command published in the GitHub repository oborchers/fractional-cto (32 stars, last pushed 5d ago), licensed MIT. It adds 21 tokens to every session and 1,030 once invoked, about $0.0001 per session on Opus 5.5. 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.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
speckit.bug.fix
Apply the remediation from a bug assessment and record what was changed.