asset-sourcing
Skill0 starsChanged 5 months agoMITView on GitHub
Provision REAL external data/assets automatically (a fetch script wired to the build, idempotent, with license and declared fallback), never placeholders. Any domain. Use it when the product needs an external source.
In this repository20 other files it gives its agents.
- docs/AGENTS.mdAGENTS.md
- adopt-groundSkill
- agent-authoringSkill
- architecture-decisionSkill
- brand-evaluation-rubricSkill
- criticSkill
- data-integritySkill
- diagrams-gateSkill
- drift-protocolSkill
- env-reproducibleSkill
- orchestrator-bridgeSkill
- project-cleanupSkill
- qa-eyesSkill
- qa-visual-criteriaSkill
- research-methodologySkill
- security-auditSkill
- setupSkill
- smoke-testSkill
- technical-documentationSkill
- visualizerSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 19
Wired into
build:all: BEFORE the build step → provisioning is NOT a manual step for the PO. - Distinctive: line 21
DECLARED fallback (P-11) + explicit error (P-06): if a download fails, a clear error (which source, what happened) and DECLARED degradation …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
SKILL: asset-sourcing — REAL data/assets, provisioned, never placeholders
Rule (P-11 with teeth): before using any EXTERNAL datum or asset the product needs, obtain the REAL source and PROVISION it automatically. If there is no source, ESCALATE; never fall back to placeholder/mock/initials/empty data without DECLARING it. Applies to ANY domain (P-10).
Scope (any domain)
Covers EVERY external asset/datum the product requires, not just code or art:
- datasets and reference tables/registries
- typefaces (fonts)
- icons and art
- models
- lookup tables / reference files
Provisioning pattern (mandatory)
- Fetch script (e.g.
scripts/fetch-assets.*) that downloads the REAL source to the exact path where the code consumes it. - Wired into
build:allBEFORE the build step → provisioning is NOT a manual step for the PO. - Idempotent: cache + verify integrity; if it already exists and is valid, do not re-download.
- DECLARED fallback (P-11) + explicit error (P-06): if a download fails, a clear error (which source, what happened) and DECLARED degradation (the missing datum is marked as such; the lookup/resolver returns an honest empty) — NEVER a placeholder that looks real, nor a silent failure.
- License attribution: record each source's license in a
NOTICEfile. - No runtime CDN dependency (P-02): the asset is bundled/cached LOCALLY.
Environment note
The agent's environment may be unable to download (network limits). The studio still AUTHORS the automation (script + build wiring); the PO's build EXECUTES it. The download is never delivered as a "manual PO task".
Tools (reference)
- UI icons: better-icons (https://github.com/better-auth/better-icons) —
npx better-icons setup; brings the real SVG LOCAL, no runtime CDN. - Other data/assets: the appropriate canonical source, downloaded at build into a resources folder and cached locally.
QA audits it: an external datum/asset delivered as a manual step, or an undeclared placeholder, is a DEFECT — not a delivery.
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
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.

