What it is
Anthropic’s reference skill for building with the Claude API and SDKs across languages — models, tool use, streaming, caching and managed agents.
Reference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration.
TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens).
SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).
Benchmark results
This skill has not been benchmarked yet. Only the static health check below has been run. Request a benchmark
Static health check
Checker 1.0.0 at commit ca45fc4: 13 pass, 2 warning, 1 fail, 5 informational. Result last changed 2026-10-29.
- Bundles scripts the agent may run
- Scripts call the network
| Check | Result | Detail |
|---|---|---|
frontmatter.descriptiondescription is present and at most 1024 characters; under 50 is a warning. | Fail | description exceeds 1024 characters |
body.tokensApproxApproximate token count of the SKILL.md body; above 5000 is a warning. | Warning | large body; consider moving detail into referenced files |
body.linesLine count of the SKILL.md body; above 500 is a warning. | Warning | 608 lines |
frontmatter.unknownKeysFrontmatter keys outside the documented set (information only). | Info | none |
files.inventoryNumber and total size of files in the skill directory (information only). | Info | 96 files, 2079868 bytes |
scripts.presentBundled scripts the agent may run (information only). | Info | 4 files
|
scripts.networkScripts that call the network (information only). | Info | shared/preserved-thinking-migration/drop_block_probe.py |
pinned.driftWhether upstream changed the skill after the pinned commit (information only). | Info | unchanged since the pinned commit |
skillmd.presentA SKILL.md file exists at the skill root. | Pass | SKILL.md found |
frontmatter.parseSKILL.md starts with YAML frontmatter that parses as a mapping. | Pass | frontmatter parsed |
frontmatter.namename is present, at most 64 characters, lowercase letters, digits and hyphens. | Pass | valid name |
frontmatter.nameMatchesDirname matches the directory that contains SKILL.md. | Pass | name matches its directory |
frontmatter.descriptionTriggerdescription says when to use the skill ("use when…", "when the user…"). | Pass | states when to use the skill |
frontmatter.descriptionVoicedescription is written in the third person and contains no XML tags. | Pass | third person, no XML |
refs.brokenLinksRelative links in SKILL.md point to files that exist. | Pass | all relative links resolve |
files.binaryNo executables or archives (.exe, .zip, .jar, …) are bundled. | Pass | no executables or archives |
scripts.destructiveScripts with destructive commands such as rm -rf, sudo or chmod 777. | Pass | none found |
secrets.scanNo strings that look like private keys or API tokens. | Pass | none found |
instructions.riskNo instructions to override prior instructions, exfiltrate secrets or bypass safety controls. | Pass | none found |
license.detectedA license is declared in frontmatter, the skill directory or the repository. | Pass | declared in frontmatter |
repo.activityThe skill path changed within 365 days of the check, and the repository is not archived. | Pass | last changed 19 days before the check |
Limits of this report
- Results describe one commit of the skill. Later versions may behave differently.
- The static check reads files; it does not show whether the skill helps on a task.
- The skill's author was not consulted before publication. Corrections are welcome. Report an error
Questions
- What does SkillFlux's data say about Claude API reference?
- As of 2026-10-29, SkillFlux's static health check of Claude API reference at commit ca45fc4 recorded 13 passes, 2 warnings and 1 failures, plus 5 informational results, across 21 checks; it has not been benchmarked yet.
- Has Claude API reference been benchmarked?
- Not yet. Only the static health check has been run. You can request a benchmark by email from the submit page.
- What license is Claude API reference under?
- The skill ships its own license file (skills/claude-api/LICENSE.txt); read it before reuse.
How to cite
SkillFlux (2026-10-29). Claude API reference health check, commit ca45fc4, checker 1.0.0. https://skillflux.app/skills/anthropics-skills-claude-api/
BibTeX
@misc{skillflux_anthropics_skills_claude_api_20261029,
author = {{SkillFlux}},
title = {Claude API reference health check (commit ca45fc4)},
year = {2026},
month = {10},
howpublished = {\url{https://skillflux.app/skills/anthropics-skills-claude-api/}},
note = {Methodology v1. Data: https://skillflux.app/data/skills/anthropics-skills-claude-api.json. Accessed 2026-10-29.},
}