Companion products for AI-assisted software work.
SaferSkills independently audited alignfirst (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
A bulleted imperative like {match} tells the agent to never reveal, disclose, or mention something to the user. Used adversarially it can instruct the agent to hide its tool calls or lie about what it did — stripping the transparency a user relies on to trust the agent.
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.
If you don't already know which protocol to use, read overview.md first.
TASK_DIR is the directory where work files related to a task are stored. Usually, we use TASK_DIR = .plans/{TICKET_ID}/ (a sub-directory of the .plans folder). If no ticket ID is known, ask the user for it.
Format: {CYCLE_LETTER}{FILE_NUMBER}-{FILE_TYPE}.md
Common file types:
spec - technical specificationplan - implementation planAAD.summary - AAD summary documentdescription - PR/MR descriptionreview - code review reportmerge.summary - merge conflicts resolution summaryExample structure:
.plans/
├── 123/
│ ├── A1-spec.md
│ ├── A2-plan.md
│ └── A3-AAD.summary.md
│ └── B1-spec.mdA if there is no existing cycle. So you just need to ask for a ticket ID if you don't have one.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.