🛠️ Skill Audit

2026-10-16 · 2 skills modified in past 48h
0
skills scored ≤4 / 10
0
merge proposals
avg 7.0 · median 7 · 2 scored · 0 unscored · 681 tokens to load

📋 Review queue lowest score first · tap "why?" for proposal

ai-bubble-tracker
7/10
Tracks AI bubble warnings and trigger events daily.
PATCH-HEADLINEPATCH-COMPACT
(top-level) · 2.1h ago · 924 words · 338 tok to load
head
5/10
goal
10/10
comp
5/10
contra
10/10
useful
10/10
unique
10/10
pub
0/10
why this score?
goal 2/2
  • body is empty after frontmatter
  • no 'when to use' / 'load this when' trigger in opening paragraph
useful 2/2
  • no runnable shell command block — agent has to improvise
pub 0/2
  • missing 'when NOT to use' — accidental dispatch risk
see proposed fix
PATCH-HEADLINE
LLM: current description doesn't precisely describe the goal
PROPOSED FIX
replace the `description:` line with: description: "Use when tracking named experts predicting an AI bubble burst. Scans Burry/Grantham/Dalio et al daily, watches for trigger events, and rebuilds a dashboard."
PATCH-COMPACT
LLM: body has restatement or prose bloat
PROPOSED FIX
tighten the body — a working tighter version: # AI Bubble Tracker A daily tracker of named experts and institutions calling for an AI bubble burst in 2027-2028, with automated watches for the triggers they care about. ## When to use - "What did the AI bubble tracker find today?" - "Did Burry / Grantham / Dalio post anything new?" - "Run the AI bubble scan" - "Rebuild the AI tracker dashboard" - "Update the AI bubble tracker predictions" ## Data files (all in `~/.hermes/workspaces/nancy/ai-bubble-tracker/`) - `data/predictions.json` — the 11 named sources + 5 triggers - `data/events.json` — append-only log of detected events - `data/signal_history.json` — per-day `{source_id, severity}` rows; drives the all-time heat count - `data/seen_ids.json` — dedup key (URL) → first-seen-timestamp; used by the daily cron to avoid re-logging the same article - `index.html` — last-rendered dashboard (dark default, light toggle) ## Scripts - `scripts/scan.py` — read-only post-processor: prints Slack-ready summary - `scripts/render.py` — r…
PATCH-PUBLISH
LLM: internal references leak — strip before public
  • missing 'when NOT to use' — accidental dispatch risk
PROPOSED FIX
replace each internal reference: line 15: ## Data files (all in <workspace>/ai-bubble-tracker/) line 25: Daily — scheduled cron job scans sources, dedups via seen_ids.json, appends to events.json + signal_history.json, and posts a summary to a configured channel. line 27: Weekly — full HTML page rebuild + static-host deploy. line 33: Canonical URL: https://nancy-ai-bubble.pages.dev (project nancy-ai-bubble, deployed via wrangler pages deploy; first deploy 2026-10-12. Cloudflare Pages slugs registered in the workspace's TOOLS.md.) line 41: Every cron reply — daily summary, weekly rebuild status, trigger-fire alert, "no new events" reply, and failure postmortems — MUST end with the canonical URL as its FINAL line. This rule is duplicated in the cron jobs.json prompts so it survives a fresh-session context (no MEMORY.md loaded).
cloudflare-pages-deploy
7/10
Deploy static sites to Cloudflare Pages.
PATCH-COMPACTPATCH-PUBLISH
(top-level) · 2.1h ago · 11525 words · 343 tok to load
head
10/10
goal
10/10
comp
0/10
contra
10/10
useful
10/10
unique
10/10
pub
0/10
why this score?
goal 2/2
  • body is empty after frontmatter
  • no 'when to use' / 'load this when' trigger in opening paragraph
contra 2/2
  • 42 'do not/never' phrases + 22 'always/must' phrases — possible direct conflicts at lines [34, 112, 128] vs [36, 42, 65]
pub 0/2
  • line 70: hardcoded ~/.hermes path — `cd /Users/morsy/.hermes/workspaces/<agent>/milestr`
  • line 181: hardcoded ~/.hermes path — ``terminal(command="cat /Users/morsy/.hermes/cron/jobs.json")` will silently truncate stdou`
  • line 194: hardcoded ~/.hermes path — `with open('/Users/morsy/.hermes/cron/jobs.json', 'rb') as f:`
see proposed fix
PATCH-COMPACT
LLM: body has restatement or prose bloat
PROPOSED FIX
tighten the body — a working tighter version: # Cloudflare Pages Deploy (this workspace) ## TL;DR **Default to existing infra before reaching for new tools.** Morsy's verbatim rejection: *"This is BS. We have Hermes."* If the existing cron + skills + cloudflare-pages-deploy pipeline can do the job (daily job, weekly HTML rebuild, deploy to `<slug>.pages.dev`), propose THAT, not a new external service (n8n, Make.com, separate cron → email bridge, etc.). Locked 2026-10-15. **Reuse an existing project before creating a new one.** Cloudflare Pages has a hard per-account project cap (free = 100, paid = 500), and the easy reflex of "one HTML file → one new project" burns through it. Before `wrangler pages project create`, run the project-reuse check below. Multi-file sister reports, dashboards, recurring cron outputs, and per-iteration artifacts from the same workflow all belong in ONE existing project with sibling files, not in N new projects. See Pitfall #27. **wrangler 4.105 auto-creates projects — one step is the default.** ```…
PATCH-PUBLISH
LLM: internal references leak — strip before public
  • line 70: hardcoded ~/.hermes path — `cd /Users/morsy/.hermes/workspaces/<agent>/milestr`
  • line 181: hardcoded ~/.hermes path — ``terminal(command="cat /Users/morsy/.hermes/cron/jobs.json")` will silently truncate stdou`
  • line 194: hardcoded ~/.hermes path — `with open('/Users/morsy/.hermes/cron/jobs.json', 'rb') as f:`
PROPOSED FIX
replace each internal reference: line 17: Cloudflare Pages has a per-account project cap (free = 100, paid = 500), and the easy reflex of "one HTML file → one new project" burns through it. line 39: Always stage deploys in a non-synced local dir (e.g. ~/.cache/pages-deploy/<name>/) before running wrangler pages deploy. line 73: cd /path/to/<your-workspace>/<agent>/milestr line 86: ACCOUNT_ID="<your-cloudf lare-account-id>" line 102: Account: your Cloudflare account email + account_id (find via wrangler whoami or Cloudflare dashboard).