Feedback triage

Skill fabioc-aloha/Alex_Skill_Mall/plugins/supervisor-fleet/feedback-triage

284 curated plugins for AI assistants across 16 categories: security, Azure, documentation, code quality, cloud infrastructure, and more. Works with GitHub Copilot. Drop into .github/skills/local/ and go.

Install

npx -y skills add fabioc-aloha/Alex_Skill_Mall --skill feedback-triage

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Categorize, deduplicate, and route fleet feedback to Edition fix, Mall change, or AlexMaster escalation

SKILL.md

3.7 KB, 899 tokens by cl100k_base, as published. Nobody here has run it

Feedback Triage

Process incoming feedback from ACT-Edition heirs. Triage classifies each item, deduplicates against open issues, and routes to the right destination.

When to Use

A new file lands in feedback/inbox/
A scheduled triage sweep runs (recommended weekly)
A pattern emerges across recent feedback that warrants escalation

The Three Routes

Every feedback item goes to exactly one of three destinations:

Route	When	Destination
Edition fix	The brain template needs a change (skill bug, instruction gap, prompt error)	`Alex_ACT_Edition` PR
Mall change	A skill/store is broken, stale, or missing from the marketplace	`Alex_Skill_Mall` PR
AlexMaster escalation	Touches ACT framework, non-ACT heirs, or fleet-wide governance	`AlexMaster/AI-Memory/feedback/`

Triage Steps

Step 1 — Categorize

Read the item and tag it with one of:

Category	Definition
`bug`	An artifact in Edition or Mall produces incorrect output
`friction`	The artifact works but is awkward, slow, or confusing
`feature-request`	A new artifact or capability is being requested
`success`	A win worth recording (informs future patterns)
`framework`	Touches ACT manifesto, tenets, or claims — escalate

Step 2 — Deduplicate

Search feedback/inbox/ and decisions/ for similar items in the last 90 days:

grep -ri "<keyword>" feedback/ decisions/

If a match exists:

Identical: link the new item to the existing thread, close the new file
Related (same root cause): merge into the existing thread, note the new symptom
Distinct: keep separate

Step 3 — Severity

Level	Definition	Action window
`critical`	Blocks fleet adoption or causes data loss	Same day
`high`	Blocks a heir or breaks a documented workflow	Within 1 week
`medium`	Friction but workaround exists	Next release
`low`	Cosmetic or nice-to-have	Backlog

Step 4 — Route

Apply the route table above. Move the file:

Edition fix → open PR against ../Alex_ACT_Edition with the verdict attached
Mall change → open PR against ../Alex_Skill_Mall
Escalation → write a stub to ../AlexMaster/AI-Memory/feedback/ (if AlexMaster is local) or open an issue (if remote)

Step 5 — Record

Append a row to decisions/curation-log.md (format below). This feeds the 60-day mechanical-vs-semantic re-evaluation per ADR-001.

Curation Log Format

| Date | Item | Category | Severity | Route | M/S | Notes |
|---|---|---|---|---|---|---|
| 2026-04-28 | mall-link-rot-01 | bug | medium | Mall change | M | 3 dead links in STORES.md |

Where M/S is Mechanical (script could detect) or Semantic (judgment required).

Anti-Patterns

Anti-pattern	Correction
Routing every item to "Edition fix" by default	Many items belong in the Mall or AlexMaster — read carefully
Skipping deduplication on busy weeks	Dedup is what prevents the inbox from becoming append-only noise
Auto-escalating ambiguous items	Escalation is a last resort, not a default. Ask first
Forgetting to record M/S classification	The 60-day re-eval depends on this signal

brain-curation-rules
skill-review — for items routed to Edition
/triage-feedback prompt