agentsclimarketplace

Official source check

Skill kreek/consult/plugin/skills/official-source-check

Use when external behavior must be checked against official sources.From its SKILL.md

Install
npx -y skills add kreek/consult --skill official-source-check

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

  • 1 stars1 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.

SKILL.md

2.5 KB, 543 tokens by cl100k_base, as published. Nobody here has run it

Official Source Check

Iron Law

MODEL MEMORY IS A HINT, NOT EVIDENCE, FOR VERSION-SENSITIVE BEHAVIOR.

When to Use

  • Implementation depends on current external framework, library, runtime, browser, SDK, cloud, or platform behavior.
  • The repo does not already prove the exact API, configuration, migration, browser behavior, or provider contract.
  • The user asks for current, official, documented, verified, or best-practice implementation.

When NOT to Use

  • Project-local logic with tests and no external API dependency.
  • Stable language syntax covered by the compiler, linter, or type checker.
  • Typos, formatting, prose-only edits, or mechanical metadata changes.
  • Emergency mitigation where the user accepts unverified risk.

Workflow

  1. Find the local version from manifests, lockfiles, imports, generated clients, schemas, config, or CI images.
  2. Check the narrow source of truth: official docs, release notes, migration guide, standards spec, or provider SDK reference.
  3. Compare that source with local conventions.
  4. Use the smallest source-compatible implementation.
  5. In the final claim, name the source checked or mark the claim unverified.

Verification

  • Local version or platform was identified, or its absence was reported.
  • The source of truth was checked for the exact pattern.
  • Deprecated or version-incompatible APIs were avoided.
  • Conflicts between source guidance and local convention were surfaced.
  • Unverified external-behavior claims were marked unverified.
  • Runtime behavior was proven when source guidance alone did not prove it.

Tripwires

TriggerDo this insteadFalse alarm
"I know this API"Check the local version and source of truth.Project-local helper with tests.
"The docs are too broad"Check the narrow API, migration, or release-note page.Offline task where user accepts unverified output.
"Existing code does it this way"Check whether the pattern is still supported before copying it.Repo policy pins an older supported pattern.
"I'll cite a blog post"Use official docs, standards, release notes, or provider references first.The project owns the library being edited.

Handoffs

  • Use workflow to choose the full skill set.
  • Use proof when source guidance must be backed by runtime behavior.
  • Use documentation when the checked source needs to be captured for maintainers.

What ships with it: 1 file

255 B alongside SKILL.md

agents/

Gives 0 of the 12 instructions most quality gates skills give in 543 tokens

Counted across 1,195 of the 2,094 authors here whose files we hold, read 2026-08-07

  • Read the output and check the exit codein 54 of 1195, across 14 files
  • Verify requirements using a line-by-line checklistin 53 of 1195, across 12 files
  • Identify the verification command proving the claimin 51 of 1195, across 12 files
  • Run the full verification commandin 50 of 1195, across 11 files
  • Verify output confirms the claimin 49 of 1195, across 12 files
  • Check version control diff after agent delegationin 46 of 1195, across 6 files
  • State claim with evidencein 44 of 1195, across 4 files
  • Run the test suitein 33 of 1195, across 26 files
  • Keep state in memory by defaultin 27 of 1195, across 6 files
  • Make prototype runnable with one commandin 26 of 1195, across 5 files
  • Produce a verification reportin 25 of 1195, across 14 files
  • Detect the package manager from lockfilesin 24 of 1195, across 5 files

Said here and by no other author read

  • treat model memory as a hint only
  • check the official source of truth
  • compare the source with local conventions
  • use the smallest source-compatible implementation
  • report missing local versions
  • avoid deprecated APIs

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,861. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.