package — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited package (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Gather audience context and shape a deliverable for a specific reader. The same information packaged differently for a technical lead vs. a CEO vs. a new hire.
Arguments provided: $ARGUMENTS
Extract from $ARGUMENTS:
If no audience provided: respond with "Usage: /package <audience> [--type <type>] [--topic \"<topic>\"]"
Search for an audience profile in order of priority:
audience-profiles/<audience>.md (project-level).claude/audience-profiles/<audience>.md (user-level)If found, extract:
If not found, build a lightweight profile from the audience name:
## Inferred Profile: {audience}
Based on the role/name provided, assuming:
- Communication style: [infer from role -- engineer=technical, executive=outcomes-focused]
- Key interests: [infer from role]
- Format: [default to structured bullet points]
[!] This is an inferred profile. Create `audience-profiles/{audience}.md` for better results.Based on --type and --topic, collect the raw material to package:
# Recent work
git log --oneline -10 2>/dev/null
# Project state
cat README.md 2>/dev/null | head -50
# Relevant files (if --topic narrows the scope)
# Search for topic-related content in the project| Type | Primary Sources |
|---|---|
| brief | Project state, recent decisions, key metrics |
| Recent conversations, action items, tone from profile | |
| report | Data, analysis, findings, recommendations |
| memo | Decision context, options, tradeoffs |
| presentation | Key points, visuals, narrative arc |
| meeting-prep | Recent context, open items, stakeholder priorities |
Apply the audience profile to the raw material:
Before finalizing, verify:
.scratch/deliverables/{audience}-{type}-{date}.md for draftsFor best results, create audience profiles at audience-profiles/<name>.md:
# Audience Profile: {Name}
## Role
{Job title or relationship}
## Communication Style
- Technical depth: [none | basic | moderate | deep]
- Jargon tolerance: [low | medium | high]
- Preferred length: [brief | standard | detailed]
- Format preference: [bullets | narrative | structured | visual]
## What They Care About
- {Priority 1}
- {Priority 2}
- {Priority 3}
## What to Avoid
- {Anti-pattern 1}
- {Anti-pattern 2}
## Language
- Primary: {language}
- Notes: {any language-specific preferences}
## Examples of What Works
- {A deliverable format or approach that landed well}/package ceo --type brief --topic "Q1 infrastructure costs"
>> Shapes a brief focused on business impact and cost, minimal technical detail
/package engineering-lead --type report --topic "database migration options"
>> Shapes a report with full technical depth, comparison tables, risk analysis
/package new-hire --type memo --topic "how our deployment pipeline works"
>> Shapes an explanatory memo assuming no prior context, defines all terms
/package investor --type email --topic "product update"
>> Shapes a concise email focused on traction metrics and milestonesThe same information fails or succeeds based on how it's packaged. A database migration plan written for an engineer works. That same document sent to a CEO fails -- not because the content is wrong, but because the packaging is wrong. This skill makes the packaging automatic.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.