Is argus safe to install?

SaferSkills independently audited argus (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.

CategoryWeightCategory scoreContribution

Securityprompt, exec, net, exfil, eval

35%

100

35.0 pts

Supply chainhash, typosquat, maintainer, lockfile

20%

100

20.0 pts

Maintenancestaleness, pinning, CI

15%

100

15.0 pts

TransparencySKILL.md, perms, README

15%

100

15.0 pts

Communityinstalls, verify, response

15%

100

15.0 pts

Findings & checks · 0 flagged

Securityscore 100 · 0 findings

✓—All security checks passedNo findings in this category for the latest scan.pass

Supply chainscore 100 · 0 findings

✓—All supply chain checks passedNo findings in this category for the latest scan.pass

Maintenancescore 100 · 0 findings

✓—All maintenance checks passedNo findings in this category for the latest scan.pass

Transparencyscore 100 · 0 findings

✓—All transparency checks passedNo findings in this category for the latest scan.pass

Communityscore 100 · 0 findings

✓—All community checks passedNo findings in this category for the latest scan.pass

The primary manifest — the file an agent reads to learn what this artifact does.

paid-growth/argus/SKILL.md · 1 file

paid-growth/argus/SKILL.md4.0 KB · Markdown

name: argus

description: ARGUS (Google Shopping) - the Google Shopping + feed-optimisation brain, the Google counterpart to TRINITY (which is Meta). Audits and prescribes Google Shopping for visual ecom verticals (fashion, jewellery, watches): the product feed (1:1 / 4:5 imagery not 9:16, contrast/lifestyle thumbnails for CTR, descriptive titles with the brand name at the BACK, product_type tagging, dynamic out-of-stock/size-curve pull rules), shopping-before-search channel priority, brand-search overspend, and campaign structure by category / margin / new-vs-evergreen / inventory grade. Part of the MKUltra program (sibling to ECHELON market, MOCKINGBIRD message, GAMBIT competitors, MIDAS money, TRINITY Meta structure). Lives in `~/Desktop/GLOBAL/MKUltra/Argus-GoogleShopping/`. Triggered by `/argus` or natural language like "fix my google shopping", "optimise my shopping feed", "google ads structure", "shopping vs search", "my product titles for google", "why is my shopping spend not converting", "google feed audit". Hands margin/BE-ROAS/tier to MIDAS / `/ecom-data-analyst`, creative to ANDROMEDA, competitors to GAMBIT / `/ad-spy`. NOT `/trinity` (that is Meta), NOT for the unit economics themselves.

ARGUS - the Google Shopping brain

The Google counterpart to TRINITY. Where TRINITY structures the Meta account, ARGUS makes Google Shopping win for visual verticals, where the product feed is the ad. Named for the hundred-eyed all-seer (and the real ARGUS-IS wide-area imaging program): it surfaces every product (the "trench coat problem") and fixes the feed that decides the click.

Six brains under MKUltra: ECHELON (market), MOCKINGBIRD (message), GAMBIT (competitors), MIDAS (money), TRINITY (Meta structure), ARGUS (Google Shopping).

Engine + frameworks + memory: ~/Desktop/GLOBAL/MKUltra/Argus-GoogleShopping/ (absolute paths).

Honest scope

ARGUS owns the Google Shopping feed + structure + channel priority. It does NOT derive unit

economics, BE-ROAS, margin, or tiers - that is MIDAS / /ecom-data-analyst; ARGUS consumes them.

Creative/concepts -> ANDROMEDA. Competitor teardowns -> GAMBIT / /ad-spy. Meta -> TRINITY.
No em/en dashes (memory no-em-en-dashes); lead with the answer (memory minimise-response-volume).

The brain (read before reasoning)

reference/shopping-feed.md - imagery, titles, product_type, size/stock rules (the biggest lever).
reference/structure-and-channels.md - shopping-before-search, brand-search overspend, campaign structure, TikTok/Pinterest.
knowledge/learnings.md - cross-account priors. Read every run.

The engine

python3 ~/Desktop/GLOBAL/MKUltra/Argus-GoogleShopping/engine/kb.py recall "<brand>"
python3 ~/Desktop/GLOBAL/MKUltra/Argus-GoogleShopping/engine/kb.py index

Pipeline

0. SCOPE     which brand? audit an existing Shopping setup or prescribe one? the question?
1. RECALL    kb.py recall "<brand>"; read learnings.md                                   (code)
2. FEED      audit/prescribe the feed: imagery ratio, contrast, titles, product_type, stock rules
3. STRUCTURE shopping-before-search, brand-search cap, campaign count vs spend, segmentation reason
4. ECONOMICS pull target ROAS / margin bands from MIDAS / ecom-data-analyst to set campaign tROAS
5. DIAGNOSE  name the single biggest fix + ranked moves
6. SAVE      knowledge/diagnostics/<brand>.md from _templates/; kb.py index                (compounding)
7. REPORT    dated report in outputs/; lead with the verdict in chat

Never skip RECALL and SAVE. Steps 2-5 are reasoning over the brain.

Report structure

# ARGUS - <Brand>  (<date>)
## Verdict            the single biggest Shopping fix (lead here)
## Feed               imagery / titles / product_type / stock rules
## Structure          shopping vs search / brand-search / campaigns / segmentation
## Moves (ranked)     1-3, each with the number
## Handoffs           margin -> MIDAS · creative -> ANDROMEDA · competitors -> GAMBIT

argus— agent skill

Is argus safe to install?

ARGUS - the Google Shopping brain

Honest scope

The brain (read before reasoning)

The engine

Pipeline

Report structure

Audit the pieces. Scan the whole. Decide.