agentleFS
mkreindelmkreindel/seo-toolkit-template/.claude/skills/triage/SKILL.md

triage

Skill1 starsChanged 6 months agoApache-2.0View on GitHub

Score every site in `sites/` by SEO opportunity and recommend the top 3 to focus investment on. Considers revenue potential, current authority, technical health, content gap, and conversion potential. Outputs a ranked list with rationale. Use when the user types `/triage` or asks "which sites should I focus on.

190 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use writing-beats, setup-matt-pocock-skills and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • LesaniSkill

    writing-beats

    Shape an article as a journey of beats, choose-your-own-adventure style. The user picks a starting beat from the raw material, you write only that beat, then offer options for where to pivot next, beat by beat, until the article reaches a natural end. Use when the user has raw material and wants to assemble it as a narrative rather than an argument.

    Also in 74 projects that use this

  • GoogleSkill

    setup-matt-pocock-skills

    Configure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.

    Also in 182 projects that use this

  • SethGammonSkill

    ascii-diagram

    Generate perfectly aligned ASCII diagrams — architecture, flow, sequence, box-and-arrow. Uses a programmatic character-grid approach so alignment is guaranteed by math, not token prediction. Includes post-render verification.

    Also in 4 projects that use this

  • simo97Skill

    migrate-to-shoehorn

    Migrate test files from `as` type assertions to @total-typescript/shoehorn. Use when user mentions shoehorn, wants to replace `as` in tests, or needs partial test data.

    Also in 74 projects that use this

  • LesaniSkill

    writing-shape

    Take a markdown file of raw material and shape it into an article through a conversational session — drafting candidate openings, growing the piece paragraph by paragraph, arguing about format (lists, tables, callouts, quotes) at each step. Use when the user has a pile of notes, fragments, or a rough draft and wants help turning it into something publishable.

    Also in 70 projects that use this

  • liufeng1976Skill

    setup-ts-deep-modules

    Wire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files. User-invoked.

    Also in 47 projects that use this

  • +1 more from Lesani

The file

We split this skill into the rules it gives an agent

10 rules. 0 are common, 1 is distinctive, and 9 are very specific to this project.

  • 0

    Common

    Most of the 179 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 9

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If the invocation contains --cron, this skill runs in cron mode (no user available).

    line 21
  2. Very specific:

    Defaults: when a choice would normally be asked, default to site-info.md / goals.md values.

    line 27
  3. Very specific:

    Audit log: ALWAYS write one line to the audit log on exit — success (shipped), escalation (escalated), idempotent skip (idempotent-skip), …

    line 29
  4. Very specific:

    Read from site-info.md → Business → "What they do" + Conversion → "What counts as a conversion" + recent revenue notes.

    line 47
  5. Very specific:

    Read from keywords.csv row count + competitor analysis in notes.md if present.

    line 73
  6. Very specific:

    Read from site-info.md → Conversion section.

    line 81
  7. Very specific:

    Read sites/{site}/goals.md.

    line 100
  8. Very specific:

    Write to triage-[YYYY-MM-DD].md at the toolkit root (gitignored if user prefers).

    line 174
  9. Distinctive:

    If most recent triage was > 90 days ago, flag this.

    line 178
  10. Very specific:

    All sites have hard pause flag (nothing to rank).

    line 195

“Common” is measured: kept by at least 6 in 10 of the 179 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/triage — Portfolio prioritization

Helps the user pick which sites in their portfolio deserve SEO investment this quarter. With 20+ managed sites and limited time, this is the highest-leverage decision.

Inputs

Optional:

  • Time horizon — quarter (default), month, year. Adjusts how forward-looking opportunities are weighted.
  • Capacity — number of sites the user can realistically focus on (default: 3).
  • Filter — optionally limit to sites matching a tag in notes.md (e.g., "client-work", "personal", "high-revenue").

Workflow

Step 0 — Cron-mode detection (if invoked with --cron)

If the invocation contains --cron, this skill runs in cron mode (no user available). Required behavior:

  1. Idempotency check: see this skill's "Routine versioning + idempotency contract" section. If today's output already exists, exit cleanly with exit: "idempotent-skip". Write one line to the audit log via scripts/lib/audit-log.mjs appendRun({ exit: "idempotent-skip", ... }).
  2. Escalation contract: any decision that would normally prompt the user (missing required file, voice anti-AI failure, schema validation failure, keyword cannibalization, etc.) MUST be escalated by writing an item to sites/{site}/_inbox/ via scripts/lib/cron-mode.mjs writeInboxItem(...). After writing, exit cleanly with exit: "escalated". Do NOT use AskUserQuestion in cron mode.
  3. Defaults: when a choice would normally be asked, default to site-info.md / goals.md values. If both are silent on the required choice, escalate per (2).
  4. Audit log: ALWAYS write one line to the audit log on exit — success (shipped), escalation (escalated), idempotent skip (idempotent-skip), or failure (failed).
  5. Backoff: at the start of every cron-mode run, call checkBackoff({ routine }) from scripts/lib/audit-log.mjs. If true, the routine has hit the 3-strike threshold — write _inbox/routine-disabled-{name}.md, run scripts/sync-schedules.mjs --pause-routine={name}, and exit.

Step 1 — Read context (Rule 1)

Load every sites/[name]/site-info.md and notes.md. If any site is missing required files, flag it but don't fail.

Step 2 — Score each site on five dimensions

For each site, compute a score 0–10 on each dimension:

Dimension 1: Revenue potential (weight: 30%)

  • 10 = real revenue source, money-keyword rankings would meaningfully grow income
  • 7 = active business, ranking gains help but aren't transformative
  • 4 = side project, lifestyle business, low monetization
  • 1 = vanity site, internal tool, no revenue intent

Read from site-info.md → Business → "What they do" + Conversion → "What counts as a conversion" + recent revenue notes.

Dimension 2: Current authority (weight: 20%)

Score based on site-info.md SEO baseline:

  • 10 = > 10K monthly organic visits, DR > 40, indexed > 500 pages
  • 7 = 1K–10K visits, DR 20–40, indexed 100–500
  • 4 = 100–1K visits, DR < 20, indexed 20–100
  • 1 = new site, < 100 visits, < 20 pages indexed

Higher-authority sites compound faster — investment pays off sooner.

Dimension 3: Technical health (weight: 15%)

From most recent /audit results in site-info.md:

  • 10 = all 8 non-negotiables pass, Lighthouse mobile ≥ 90, no Tier 1 gaps on top pages
  • 7 = minor fixes needed (1–2 non-negotiables fail or floor < score < 90)
  • 4 = significant gaps (CSR Lovable site without prerender, missing schema, Lighthouse < 70)
  • 1 = fundamentally broken (no sitemap, blocked by robots, manual action, mostly thin content)

Critical filter: a site scoring 1 here gets prioritized for technical fixes regardless of other dimensions, because content investment is wasted on a broken foundation.

Dimension 4: Content gap / opportunity (weight: 20%)

  • 10 = strong keyword research done (high-value keywords identified), competitors rank for terms this site doesn't, clear content path forward
  • 7 = some opportunity, partial keyword research
  • 4 = saturated niche or thin opportunity
  • 1 = no opportunity (already #1 for everything relevant, or topic is dead)

Read from keywords.csv row count + competitor analysis in notes.md if present.

Dimension 5: Conversion potential (weight: 15%)

  • 10 = clear conversion path documented (CTAs, form, booking system) + tracking working
  • 7 = partial setup, conversions counted but not optimized
  • 4 = weak CTAs, no tracking, no clear path
  • 1 = no conversion infrastructure (rankings would generate traffic but no revenue capture)

Read from site-info.md → Conversion section.

Step 3 — Compute weighted score

Total = (Revenue × 0.30) + (Authority × 0.20) + (Technical × 0.15) + (ContentGap × 0.20) + (Conversion × 0.15)

Score range: 1.0 to 10.0.

Step 4 — Apply filters and special rules

  • Critical technical issue (Dimension 3 = 1): site jumps to top of "fix first" list, separate from main ranking.
  • Hard pause flag in site-info.md: site removed from ranking entirely.
  • Filter argument: restrict ranking to sites matching the tag.
  • Architecture gap (SEO_GUIDE.md Section 2.6): any site with service-business: true AND > 3 rows in service-keywords.csv AND no architecture.md gets an inline flag in its rationale: ⚠️ Missing transactional architecture — required before next /service run. This doesn't change the score but surfaces the planning blocker so it's visible at the recommendation stage.

Step 4.5 — Compute goal-progress per site (T2-J)

For each site in sites/*:

  1. Read sites/{site}/goals.md. Parse the ## Success metrics section into a list of {metric, target} pairs.
  2. For each metric, attempt to measure the current value:
    • "Blog count: N+ new posts" → count entries in used-keywords.md shipped since quarter-start (parse the "Shipped — blog posts" section).
    • "Primary keyword rank: top-N for X" → if SEMrush API configured, query Position Tracking; otherwise mark cannot-measure-automatically.
    • "Backlinks: ≥N total in Referring Domains" → if SEMrush API configured, query backlinks endpoint.
    • "GSC indexed pages: ≥N" → query GSC API coverage endpoint (uses GSC_OAUTH_REFRESH_TOKEN).
    • "Sitemap grows from N → M URLs" → diff live sitemap.xml count against the target.
    • Free-text metrics (e.g., "GBP reviews: ≥5 by end of month") → mark cannot-measure-automatically; user fills in monthly.
  3. Compute "% to target" where possible: current / target * 100, clamped to 100% for the "exceeded" case.
  4. Assign visual indicator:
    • ✅ on track (>= 80%)
    • ⚠️ behind (50–79%)
    • ⏸️ off-track (< 50%)
    • 🔵 exceeded (> 100%)
    • ❓ cannot-measure-automatically (free-text or API unavailable)

Step 5 — Generate ranked list + rationale

Output format:

# Portfolio triage — [date]

**Capacity:** [N] sites | **Horizon:** [quarter/month/year]

## Fix-first (technical issues blocking content investment)

| Site | Score | Critical issue |
|------|-------|---------------|
| [name] | 1.0 (Dim 3) | [e.g., CSR Lovable site, no sitemap] |

## Top [N] for SEO investment this [horizon]

### 1. [site-name] — total score: [X.X]
- **Revenue potential:** [score]/10 — [one sentence rationale]
- **Authority:** [score]/10 — [details]
- **Technical:** [score]/10 — [recent audit summary]
- **Content gap:** [score]/10 — [keyword opportunity summary]
- **Conversion:** [score]/10 — [conversion infrastructure status]
- **Recommended next 3 actions:** [bulleted, specific — if `Architecture gap` flag is present, "Create `architecture.md` from `templates/architecture.md`" is action #1]

### 2. [site-name] — total score: [X.X]
[same structure]

### 3. [site-name] — total score: [X.X]
[same structure]

## Park (don't invest this [horizon])

[remaining sites, listed with one-sentence reason]

## Suggested cadence across the top 3

- **Site 1 (highest score):** [N posts/week + service pages + off-page actions]
- **Site 2:** [allocation]
- **Site 3:** [allocation]

Aggregate publishing across the 3 should fit the user's realistic weekly capacity.

## Goal-progress per site (T2-J)

For each site, list each goals.md success metric with current value and indicator:

### [site-name]

- ✅ Blog count: 5 / 8 posts (63%)
- ⚠️ Primary keyword rank: SEMrush API rate-limited this run (recheck next cycle)
- ⏸️ Backlinks: 3 / 7 (43%) — accelerate before quarter-end
- 🔵 GSC indexed pages: 18 / 17 (106% — exceeded)
- ❓ GBP reviews: 2 / 5 (free-text metric; manual verification)

Step 6 — Save results

Write to triage-[YYYY-MM-DD].md at the toolkit root (gitignored if user prefers). Don't overwrite previous triages — keep a history for trend analysis.

Step 7 — Quarterly re-run reminder

If most recent triage was > 90 days ago, flag this. If ≤ 90 days, note "Last triage: [date] — re-run only if portfolio composition changed."

Routine versioning + idempotency contract

This skill participates in cruise-control via --cron mode. When invoked by cron:

  1. Stamp routine_version in every output (the _triage/YYYY-MM.md cross-site summary frontmatter, per-site notes.md triage-rank entry, the audit log line via scripts/lib/audit-log.mjs). Current routine_version: 1.0.
  2. Idempotency: this skill MUST be safe to run twice in the same month without producing duplicate work. Implementation: check whether _triage/{YYYY-MM}.md already exists for the current month; if so, exit cleanly with exit: "idempotent-skip". The monthly summary is meant to be generated once on the 1st of the month.
  3. Goal-progress computation (T2-J from spec): the cron-mode run reads each site's goals.md and computes "% to target" per success metric, integrated into the per-site recommendation in the triage output.

Reference: docs/specs/2026-05-16-agents-cruise-control-design.md § Operational hardening O3 + § T2-J.

Refusal conditions

  • sites/ folder is empty (no managed sites).
  • All sites have hard pause flag (nothing to rank).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.