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 agentmods add skills/lilmgenius/polysona/qanpx skills add LilMGenius/polysona --skill qagit clone --depth 1 https://github.com/LilMGenius/polysonaWhat 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 | $0.00014 | $0.00253 |
| Opus 5 | $0.00007 | $0.00127 |
| Sonnet 5 | $0.00003 | $0.00051 |
| Haiku 4.5 | $0.00001 | $0.00025 |
Grade A, and why
qa scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
!ACTIVE=$(cat personas/_active.md 2>/dev/null || echo "default"); LATEST_DRAFT=$(ls content/drafts/*.md 2>/dev/null | sort | tail -n 1); cat "personas/$ACTIVE/accounts.md" 2>/dev/null || echo "No accounts. Run /interview first."; echo "---"; if [ -n "$LATEST_DRAFT" ]; then cat "$LATEST_DRAFT"; else echo "No drafts found. Run /content first."; fi
QA Skill Protocol
- Run in
context: forkfor isolated evaluation. - Trigger
virtual-followerto execute the QA pipeline. - Return TOP 5 recommendations with scores and concise rationale.
- Use the latest saved file in
content/drafts/as the default evaluation source when no explicit draft is provided. - Save the QA report to
content/qa/before answering. - Re-read the saved QA report and include the confirmed saved path in the response.
- If no draft file exists, report that QA is blocked instead of inventing an evaluation.
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.
- 2d ago First seen · 19 lines · 14 tokens per session scan A 8ed15517afe0
qa is a skill published in the GitHub repository LilMGenius/polysona (160 stars, last pushed 13d ago), licensed MIT. It adds 14 tokens to every session and 253 once invoked, about $0.0001 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
hive.linkedin-automation
Read before automating LinkedIn with browser tools. LinkedIn combines shadow DOM (#interop-outlet), strict Trusted Types CSP that silently drops innerHTML, Lexical composer, native beforeunload dialogs that hang the bridge, and aggressive spam filters — each has bitten us at least once. Verified flows for profile…
hive.x-automation
Read before automating X / Twitter with browser tools. Verified flows for post, reply, delete, search-and-engage, plus the Draft.js compose quirks that silently disable the send button. Includes the daily-reply and job-market-reply playbooks. Requires hive.browser-automation for the underlying screenshot + coordinate…
content-planner
WeChat Official Account topic planning and content calendar management. Based on WeChat article search and trending analysis, generates differentiated topic recommendations and outputs structured content calendars. Activated when users mention "topic", "planning", "content calendar", "trending", or "what to write next…
launch-graphics
Generate agentOS marketing/launch graphics (launch & social heroes, code-snippet cards) from the committed generators in docs-internal/graphics. Use when asked to make, update, or render an agentOS launch image, social image, "agentOS Apps" graphic, or technical code-snippet image.
tinyfish-social-listening
Monitor brand mentions, sentiment, and industry chatter across the web using TinyFish Search and Fetch. Use when a user asks to track what people are saying about a brand, product, or topic, run social listening, or get a pulse on public perception. If the user already named the target, proceed directly. Default time…
competitive-analysis
Competitive landscape analysis: positioning, scorecards, moat assessment, market share trends.