Λ-Compression — 90% - 98% Lossless Reasoning Compression
v2.0.0Physics-based lossless compression for AI output — prose AND structured data. Strips 60-98% of tokens with zero information loss. Prose mode compresses reaso...
⭐ 0· 99·0 current·0 all-time
byShadow Rose@theshadowrose
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
OpenClaw
Benign
high confidencePurpose & Capability
Name/description claim (lossless compression of AI prose and structured data) matches the skill contents: all required materials (the decoder/spec) are bundled as reference files and no external credentials, binaries, or installs are requested. Nothing requested appears unrelated to the stated purpose.
Instruction Scope
SKILL.md directs the agent to load the included references file and apply a multi-step compression/decompression protocol. The instructions do not access external endpoints, system files, or environment variables. Important caveat: correct decompression depends on the receiver sharing the decoder and conventions — using this on outputs consumed by humans or systems without the decoder will produce unreadable shorthand. The SKILL.md itself acknowledges verification and cross-model penalties.
Install Mechanism
There is no install spec and no code files executed at runtime; the skill is instruction-only and includes its entire decoder/spec as reference files. This is the lowest-risk install profile.
Credentials
The skill requests no environment variables, credentials, or config paths. No sensitive data or unrelated credentials are required.
Persistence & Privilege
The skill is not always-enabled, does not modify other skills or system settings, and contains no installation routines. It does not request elevated privileges or persistent presence.
Assessment
This skill is internally coherent and self-contained, but treat its aggressive 'lossless' claims skeptically until you verify them in your use case. Before adopting: (1) test compress/decompress round-trips on representative inputs and across any models/systems that will consume the compressed output (cross-model penalty can cause gaps); (2) do not enable it as a global default for human-facing replies or creative prose — compressed output requires the decoder and conventions to be present; (3) validate that anything the method classifies as 'derived' really is reconstructible for your consumers (sensitive or novel fields mistakenly classified as derived could be lost); and (4) note the author/contact is minimal — if you plan production use, ask for reproducible test vectors and independent verification of the losslessness claims.Like a lobster shell, security has layers — review code before you run it.
compressionvk977xxzntgxk4x95ecekeh17ax84pmevcost-reductionvk977xxzntgxk4x95ecekeh17ax84pmevevidence-classvk977xxzntgxk4x95ecekeh17ax84pmevlatestvk970hyyt1g5ymmz107qn3wbjg584qeptlosslessvk977xxzntgxk4x95ecekeh17ax84pmevphysicsvk977xxzntgxk4x95ecekeh17ax84pmevreasoningvk977xxzntgxk4x95ecekeh17ax84pmevtoken-reductionvk977xxzntgxk4x95ecekeh17ax84pmev
License
MIT-0
Free to use, modify, and redistribute. No attribution required.