CtrlK
BlogDocsLog inGet started
Tessl Logo

meta-ad-builder

Publish finished creatives as live Meta (Facebook/Instagram) ads via the Meta Marketing API, plus research and ad-copy support. Uploads an image or video, builds a multi-variant TEXT_LIQUIDITY creative, and creates a PAUSED ad in an existing ad set. Also pulls top-performing ads (ranked by ROAS) and competitor ads from the Ad Library to inform copy. Use when the user asks to deploy / publish / launch a creative as a Meta or Facebook ad, build a Meta ad, push a video or image into an ad set, pull their top ads, or research competitor ads. Not for generating creative (use the image/video skills) and not for writing AdTable/Airtable rows (use adtable-light).

74

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

85%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A tightly written, highly actionable workflow document with excellent sequencing and safety validation for a money-adjacent operation. Its main weakness is that the progressive-disclosure layer is broken in practice: every referenced detail file cited in the read order is absent from the bundle, leaving the copy-phase JSON format and API details unreachable.

Suggestions

Ship the referenced files (prompting/copy-guide.md, reference/deploy-patterns.md, reference/meta-api-cheatsheet.md) or fix the paths so the read-order links resolve — currently all three are missing from the bundle.

Include a minimal inline example of the copy.json shape (one body, one title, one description with required keys) so Phase 2 is executable even without copy-guide.md.

Clarify the directory layout: the body uses both 'prompting/' and 'reference/' while bundle conventions here use 'references/' — standardize so referenced paths are predictable.

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence — no concept explanations, just terse sections like "Run `bash scripts/check-meta-env.sh` to verify credentials before anything else". The PAUSED rule appearing in Workflow, Safety rules, and Cost note is deliberate emphasis on a money-adjacent risk rather than padding, so it fits the 5 anchor over the 4 ('could be trimmed') anchor.

5 / 5

Actionability

Deploy and research commands are fully concrete and copy-paste ready (e.g. the complete `deploy-ad.py --dry-run` and `pull-top-ads.py --date-preset last_30d --min-spend 100 --limit 15` invocations), but Phase 2 defers the essential `--copy-file` JSON shape to `prompting/copy-guide.md`, which is not present in the bundle — a real executable gap that keeps this below the 5 anchor.

4 / 5

Workflow Clarity

Phases are clearly sequenced (env check → optional research → copy → deploy) with a decision tree mapping user intents to phases, and the money-adjacent deploy has explicit validation: "Always dry-run first... Review the payload with the user, then deploy for real", plus error-handling notes (video polling, OAuthException retry) — matching the 5 anchor with explicit checkpoints and feedback loops.

5 / 5

Progressive Disclosure

The structure itself is well-signaled (an explicit read order with "Consult as needed; don't read end-to-end", one level deep), but all three referenced files — `prompting/copy-guide.md`, `reference/deploy-patterns.md`, `reference/meta-api-cheatsheet.md` — are missing from the bundle, so navigation breaks and critical content is unreachable. That is more than the 'minor organization gaps' of the 4 anchor, fitting the 3 anchor for incomplete reference organization.

3 / 5

Total

17

/

20

Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

An exemplary description: concrete third-person capability statements across all three phases, natural trigger phrases with synonyms, an explicit 'Use when' clause, and clear exclusions that demarcate it from sibling skills. No vague language or over-claims.

DimensionReasoningScore

Specificity

Lists multiple concrete actions covering every phase — "Uploads an image or video, builds a multi-variant TEXT_LIQUIDITY creative, and creates a PAUSED ad in an existing ad set" plus "pulls top-performing ads (ranked by ROAS) and competitor ads from the Ad Library" — with no coverage gaps, matching the comprehensive anchor rather than the 'minor gaps' anchor at 4.

5 / 5

Completeness

Explicitly answers both what (detailed capability list across research, copy, and deploy) and when ("Use when the user asks to deploy / publish / launch...") with concrete trigger phrases, plus explicit exclusions — fully matches the 5 anchor; 'when' is explicit, not just present as at 4.

5 / 5

Trigger Term Quality

Trigger phrases are natural user language with synonym coverage: "deploy / publish / launch a creative as a Meta or Facebook ad", "push a video or image into an ad set", "pull their top ads", "research competitor ads" — comprehensive including Meta/Facebook variants, fitting the 5 anchor.

5 / 5

Distinctiveness Conflict Risk

Clear niche (direct Meta Marketing API deployment) with affirmative disambiguation — "Not for generating creative (use the image/video skills) and not for writing AdTable/Airtable rows (use adtable-light)" — minimizing conflict risk with adjacent skills, matching the 5 anchor.

5 / 5

Total

20

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 5 missing

Warning

Total

15

/

16

Passed

Repository
krusemediallc/arcads-claude-code
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.