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 owainlewis/agent-skills --skill compressgit clone --depth 1 https://github.com/owainlewis/agent-skillsWrote 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/owainlewis/agent-skills/compress)<a href="https://agentmods.dev/skills/owainlewis/agent-skills/compress"><img src="https://agentmods.dev/badge/skills/owainlewis/agent-skills/compress.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 | $0.00062 | $0.00492 |
| Opus 5 | $0.00031 | $0.00246 |
| Sonnet 5 | $0.00012 | $0.00098 |
| Haiku 4.5 | $0.00006 | $0.00049 |
Grade A, and why
compress 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 8d 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 — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Compress
Shorten the input without losing meaning.
Use this for information and instructions, not ordinary prose polishing. Use deslop when voice and writing quality are the main problem.
Process
- Identify the purpose and intended reader or agent.
- Extract the required content: facts, actions, inputs, outputs, constraints, defaults, examples, edge cases, checks, and stop conditions.
- Remove repetition, throat-clearing, backstory, generic advice, motivational language, and structure that does not change understanding or behaviour.
- Replace vague phrases with a concrete instruction only when the source supports it. Otherwise cut them.
- Merge overlapping rules and place each fact once, where the reader needs it.
- Rewrite in plain words and the smallest structure that preserves the logic.
- Compare the result with the source. Restore anything whose removal changes the likely interpretation or outcome.
Keep
- facts, numbers, names, paths, commands, schemas, and citations
- explicit scope, priorities, and exceptions
- requirements created by past failures
- defaults that resolve real ambiguity
- examples that define a format or expose an edge case
- verification, failure behaviour, and stop conditions
- useful voice when compressing a brain dump or note
Cut
- repeated goals, summaries, and conclusions
- obvious preambles and narration about the document
- baseline advice such as "be thorough" or "use good judgement"
- adjectives such as "robust," "polished," and "production-ready" without a test
- hedging that does not change confidence or responsibility
- examples that merely repeat the rule
- headings, bullets, or steps that add shape but no meaning
- synonyms for a term already named clearly
Test every cut
Ask: if this disappears, could a capable reader act differently or lose a fact, limit, decision, warning, or piece of voice?
If no, cut it. If yes, keep it or rewrite it more clearly.
Return only the compressed result unless the user asks for a summary of changes.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 8d ago First seen · 50 lines · 62 tokens per session scan A a495f45bb1e1
compress is a skill published in the GitHub repository owainlewis/agent-skills (44 stars, last pushed 17d ago), licensed MIT. It adds 62 tokens to every session and 492 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
recipe-create-meet-space
Create a Google Meet meeting space and share the join link.
x-tweet-by-conversation
Collects every tweet in an X (Twitter) conversation thread given a conversation id (root tweet id) — the focal tweet plus all replies, sub-replies, and quote chains — and returns normalized per-tweet data with text, author, engagement counts, media, hashtags, mentions, inreplyto mapping, and cursor for pagination. Use…
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
atmos-config
Atmos root configuration: atmos.yaml discovery, precedence, deep merging, basepath, imports, minimal bootstrap, and routing to narrower Atmos skills.
auditing-azure-active-directory-configuration
Auditing Microsoft Entra ID (Azure Active Directory) configuration to identify risky authentication policies, overly permissive role assignments, stale accounts, conditional access gaps, and guest user risks using AzureAD PowerShell, Microsoft Graph API, and ScoutSuite.
detecting-privilege-escalation-in-kubernetes-pods
Detect and prevent privilege escalation in Kubernetes pods by monitoring security contexts, capabilities, and syscall patterns with Falco and OPA policies.