hashlips/md-compressor
skills/md-compressor/SKILL.md
Compresses Markdown to minimal token form while preserving facts and agent-executable intent; duplicates sources first and compresses skill packages (SKILL.md plus reference MD). Use when reducing context cost in docs, skills, or prompts at light, medium, heavy, or extreme compression.
$ install --global
skillsauthnpx skillsauth add hashlips/agent-skills md-compressorInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: May 24, 2026, 2:32 AM101.0s9 files scanned
SKILL.md
- name:
- md-compressor
- description:
- Compresses Markdown to minimal token form while preserving facts and agent-executable intent; duplicates sources first and compresses skill packages (SKILL.md plus reference MD). Use when reducing context cost in docs, skills, or prompts at light, medium, heavy, or extreme compression.
MD Compressor
Shrink Markdown to the smallest workable form for agents. Never edit originals — duplicate first, compress the copy only.
End Goal
The machine must still know what to do after compression — at every level, including extreme.
- Always preserve: actionable requirements, constraints, steps, decisions, numbers, commands, paths (per level rules), and skill triggers (
name, usabledescription). - May be reduced as levels increase: tone, encouragement, examples, edge-case color, and duplicate explanations — some context and “feel” can be lost; that is an expected tradeoff.
- Never acceptable: an agent reading only the compressed copy cannot determine what action to take, what order to do it in, or what must not be violated.
- If executability is at risk: keep the longer form, revert that fragment, or step down one compression level for that line or file.
Heavier compression = fewer tokens, not a different skill. When reporting results, note that heavy / extreme copies may omit nuance but must remain agent-executable.
Non-Negotiable Safety
- Duplicate before compress — copy the target skill folder or MD file; write all compressed output into the copy.
- Source stays intact — do not save compression edits to the original path.
- Confirm target — state source path, copy path, and compression level before editing.
Core Rules
- Facts are immutable — requirements, constraints, numbers, paths, commands, and decisions stay exact.
- Level is explicit — default
lightunless the user asks formedium,heavy, orextreme. - Skill = whole package — compress
SKILL.mdand every.mdunderreferences/in the duplicated folder; copy non-text assets unchanged. - Text only — compress Markdown and plain text; skip binaries, images, and non-text assets.
- No semantic drift — if shortening risks changing meaning or what the agent should do, keep the longer form.
- Agent-executable over minimal — prefer tokens saved only when the compressed copy still answers: what to do, in what order, under what constraints.
- Valid skill output — preserve YAML frontmatter keys (
name,description); keepdescriptiontrigger-clear even when shortened.
Compression Levels
| Level | Goal | Human-readable | | --- | --- | --- | | light | Tight sentences; keep structure and punctuation | Yes | | medium | Light + abbreviations, symbol swaps, denser lists | Mostly | | heavy | Minimum tokens; machine-parseable over readable | Optional | | extreme | Heavy + whitespace strip + short ref filenames and links | No |
Level definitions: references/compression-levels.md.
Workflow
- Intake — single file, folder, or skill package; pick
light,medium,heavy, orextreme. - Duplicate — references/safe-copy-workflow.md.
- Scope files — skill packages: references/skill-package-handling.md.
- Compress — per-file pass: references/compression-workflow.md.
- QA — references/quality-checklist.md.
Output Contract
- First line of each compressed file body:
Compression: light|medium|heavy|extreme(after frontmatter when present). - light / medium / heavy: output remains valid Markdown (heavy may be dense but parseable).
- extreme: whitespace-stripped; may be a few very thin lines — facts and literals intact; agent-parseable over human-readable.
- extreme (skill packages): may rename
references/*.mdto short names and rewrite all internal links; see references/extreme-path-shortening.md. - Preserve heading hierarchy unless flattening is lossless (light/medium); heavy/extreme may collapse when intent is unchanged.
- Keep external URLs, CLI paths, and fenced code literals intact; internal ref paths may shorten at extreme only.
- Add a one-line abbreviation legend at top only when heavy/extreme uses non-map shorthand (omit legend at extreme if it costs more tokens than it saves).
Reference Index
- Duplicate-first safety and naming: references/safe-copy-workflow.md
- Light / medium / heavy / extreme behavior: references/compression-levels.md
- Skill folder scope and frontmatter: references/skill-package-handling.md
- Per-file compression order: references/compression-workflow.md
- Canonical abbreviations: references/abbreviation-map.md
- Symbol and punctuation by level: references/symbol-and-punctuation-policy.md
- Extreme path shortening: references/extreme-path-shortening.md
- Final QA: references/quality-checklist.md
When To Use This Skill
- Shrinking verbose Markdown or full skill packages for lower context cost.
- Testing how small a skill can get while staying agent-usable.
- Preparing dense handoff or archive copies without touching source skills.
Related Skills
hashlips/project-graph
development
VerifiedTrustedCommunity
Generates an interactive HTML map of any coding project showing all files, their connections, and entry-point flows, enriched by a living AI knowledge file that grows as the project is explored. Use when asked to generate a project graph, visualize project structure, map file connections or flows, get a high-level overview of a codebase, or answer questions about a project that already has a generated graph (technologies, services, how a flow works).
16SKILL.mdUpdated Jun 12, 2026
hashlips/skill-scan
development
VerifiedTrustedCommunity
Static security audit of an Agent Skill package (untrusted text only)—safe to run, data-exfil and hidden-action risks, est_tokens, and a verdict. Use only when the user explicitly asks to scan, security-scan, or sanity-check a skill.
12SKILL.mdUpdated May 26, 2026
hashlips/token-estimator
development
VerifiedTrustedCommunity
Gives a consistent rough token-size estimate for a file or folder of text using fixed integer rules. Use when you need repeatable ballpark token counts for Markdown or plain text without calling a provider API.
12SKILL.mdUpdated May 24, 2026
hashlips/copy-sanitizer
data-ai
VerifiedTrustedCommunity
Naturalizes AI-assisted copy by removing statistical fingerprints while preserving the author's tone, meaning, and intent. Use when drafts sound polished but machine-smooth, when glued hyphen words need plain phrasing, or when any body of text needs a light pass to read more naturally.
11SKILL.mdUpdated May 17, 2026