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 agentmods add commands/ofershap/prompt-compression/compress-promptgit clone --depth 1 https://github.com/ofershap/prompt-compressionWrote 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/ofershap/prompt-compression/compress-prompt)<a href="https://agentmods.dev/commands/ofershap/prompt-compression/compress-prompt"><img src="https://agentmods.dev/badge/commands/ofershap/prompt-compression/compress-prompt.svg" alt="Measured on agentmods" 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 | $0.00016 | $0.00434 |
| Opus 5 | $0.00008 | $0.00217 |
| Sonnet 5 | $0.00003 | $0.00087 |
| Haiku 4.5 | $0.00002 | $0.00043 |
Grade A, and why
compress-prompt 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.
What it actually says
Compress the provided content into a compact, token-efficient format suitable for AGENTS.md or CLAUDE.md.
Use the prompt-compression skill for compression techniques and output formats.
When a file or content is provided
- Identify content type: docs index, coding standards, API reference, config/schema, or mixed
- Extract actionable content: strip prose, motivation, history, redundant examples
- Apply compression:
- Pipe-delimited index for file trees
- Single-line directives for rules and standards
- Abbreviated keys for config and schema
- Brace expansion for grouped items
- Add retrieval directive if output references readable files
- Show before/after token count (estimate: 1 token per 4 characters)
- Present the compressed output in a fenced code block ready to paste into AGENTS.md
When no file or content is provided
Guide the user through the process:
-
Ask what they want to compress:
- "What content do you want to compress for AGENTS.md? Options:"
- a) Framework docs - point me to the docs directory or framework name
- b) Project coding standards - paste or point me to your guidelines
- c) API reference - share your API docs or OpenAPI spec
- d) Existing AGENTS.md - I'll audit and compress what you have
- e) Other - describe what you want to compress
-
Once the user provides content, proceed with the compression steps above.
Quality checks
After compression, verify:
- Every file or section from the original is still reachable in the compressed version
- No actionable information was lost
- Token reduction is at least 60%
- Output is valid markdown that can be pasted directly into AGENTS.md
- Retrieval-led reasoning directive is included (if referencing files)
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 · 47 lines · 16 tokens per session scan A 6c6910cdbf1a
compress-prompt is a command published in the GitHub repository ofershap/prompt-compression (1 stars, last pushed 6mo ago), licensed MIT. It adds 16 tokens to every session and 434 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.
Other commands, from other repositories
ask-openrouter
Execute the canonical workflow: .agent/workflows/ask-openrouter.md.
prompt-history
Manage history of created and optimized prompts.
prompt
Create or optimize prompts using best practices.
prompt
System instructions for writing effective prompts. Apply when generating commands, skills, agents, or any LLM instructions.
prompt-show
Display full details of a saved prompt by ID.
eval
Evaluate and improve one healthcare agent's system prompt. Run up to 5 iterations of: prepare fixed questions -> answer -> judge -> improve -> re-score -> commit if better.