Instrumenta CLI ad codicem scribendum

Instrumentum codicis tribus statutis eget ut cum Shannon operetur: URL basi API, clave tua API et id modelli Shannon. Haec pagina statuta exacta pro Claude Code et Codex CLI dat, explicat quomodo modellum codicis shannon-coder-1 computetur, et errores quos haec instrumenta inveniunt enumerat.

Quo endpoint quodque instrumentum utatur

Omnia tria endpoints eandem clavem API accipiunt, ut Authorization: Bearer vel ut x-api-key. Headers anthropic-version, anthropic-beta et openai-beta quos haec instrumenta mittunt accipiuntur.

Modellum Shannon elige

Quaeque petitio id modelli nominare debet quod Shannon publicat. Instrumentum codicis nomen modelli sui defaultis mittit nisi aliter ei dicis, et nomen quod Shannon non publicat cum 400 et nuntio unknown model: <name> respondetur. Nullum aliud nomen ad modellum Shannon refertur, itaque modellum in instrumento ante primam petitionem pone.

GET /v1/models omnia ids cum fenestra contextus enumerat. Clavem non requirit. Modella et pretia

Instrumenta codicis modellum system prompt longo dirigunt. System prompt a shannon-coder-1, shannon-1.6-lite, shannon-1.6-pro, shannon-2-lite, shannon-2-pro et modellis open-weight hospitatis applicatur, itaque unum ex his pro instrumento codicis elige.

Claude Code

Claude Code connexionem suam e variabilibus ambitus legit. Eas in shell quod eum incipit pone, vel in blocum env fasciculi ~/.claude/settings.json ut pro omni sessione serventur.

# Install Claude Code
npm install -g @anthropic-ai/claude-code

# Where to send requests, and your Shannon API key
export ANTHROPIC_BASE_URL=https://api.shannon-ai.com
export ANTHROPIC_AUTH_TOKEN=YOUR_API_KEY

# Shannon model ids: the session model, and the model behind each built-in name
export ANTHROPIC_MODEL=shannon-coder-1
export ANTHROPIC_DEFAULT_OPUS_MODEL=shannon-coder-1
export ANTHROPIC_DEFAULT_SONNET_MODEL=shannon-coder-1
export ANTHROPIC_DEFAULT_HAIKU_MODEL=shannon-1.6-lite

# The context window of the session model
export CLAUDE_CODE_MAX_CONTEXT_TOKENS=131072

claude

{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.shannon-ai.com",
    "ANTHROPIC_AUTH_TOKEN": "YOUR_API_KEY",
    "ANTHROPIC_MODEL": "shannon-coder-1",
    "ANTHROPIC_DEFAULT_OPUS_MODEL": "shannon-coder-1",
    "ANTHROPIC_DEFAULT_SONNET_MODEL": "shannon-coder-1",
    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "shannon-1.6-lite",
    "CLAUDE_CODE_MAX_CONTEXT_TOKENS": "131072"
  }
}

• Omnes tres variabiles ANTHROPIC_DEFAULT_…_MODEL pone. Subagentes et quaedam mandata modellum nomine suo inbuilt petunt, et quodque nomen in id Shannon resolvi debet.
• Exemplum petitiones in secundo plano ad shannon-1.6-lite dirigit, quod in tokens computatur. Tituli sessionum et similes petitiones tum vocationibus Coder non utuntur.
• Claude Code POST /v1/messages/count_tokens de magnitudine contextus rogat. Endpoint pro modellis open-weight hospitatis computat. Cum id Shannon cum 400 respondet; Claude Code computationem tokenorum ut facultativam tractat et magnitudinem ipse aestimat. Computatio tokenum

Ut inscriptionem, clavem et id modelli inspicias antequam instrumentum incipis, unam petitionem manu mitte:

curl https://api.shannon-ai.com/v1/messages \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "shannon-1.6-lite",
    "max_tokens": 16,
    "messages": [{"role": "user", "content": "Say OK."}]
  }'

Codex CLI

Codex CLI cum API Responses colloquitur. Shannon ut provisorem modellorum in ~/.codex/config.toml describe, ibi id modelli Shannon elige, et clavem tuam API in variabilem ambitus quam provisor nominat pone.

model = "shannon-coder-1"
model_provider = "shannon"
model_context_window = 131072

[model_providers.shannon]
name = "Shannon"
base_url = "https://api.shannon-ai.com/v1"
env_key = "SHANNON_API_KEY"
wire_api = "responses"

# Install Codex CLI
npm install -g @openai/codex

# Your Shannon API key, under the name given as env_key
export SHANNON_API_KEY=YOUR_API_KEY

# Start it; --model selects another Shannon id for one run
codex
codex --model shannon-1.6-pro

• /v1/responses nullum colloquium in servere servat: quaeque petitio colloquium in input fert. previous_response_id propter compatibilitatem accipitur.
• Instrumenta functionum (type: "function") modello in /v1/responses offeruntur.
• In /v1/responses, shannon-coder-1 in tokens de saldo tuo computatur, non in vocationibus Coder.

Eadem inspectio pro endpoint Responses:

curl https://api.shannon-ai.com/v1/responses \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "shannon-1.6-lite",
    "input": "Say OK."
  }'

Modellum codicis et quomodo computetur

shannon-coder-1 est modellum Shannon pro agentibus codicis. GET /v1/models id cum fenestra contextus 131,072 tokenorum, vocationibus instrumentorum, ratiocinatione et input textus nuntiat. Duobus modis computatur, secundum endpoint quo instrumentum utitur.

Vocatio est una petitio, cuiuscumque magnitudinis. Computatur cum petitio accipitur, antequam modellum respondet. Sessio agentis unam petitionem pro quoque gradu quem facit mittit, itaque negotium longum multas vocationes consumit. Quota a plano tuo pendet et quaque 4 horis de integro incipit, hora 00:00, 04:00, 08:00, 12:00, 16:00 et 20:00 UTC.

Pagina planorum numeros praesentes ostendit. Plana

Cum vocationes fenestrae consumptae sunt, petitio cum 429 respondetur. In /v1/messages idem error in forma Anthropic venit.

{
  "error": {
    "type": "rate_limit_error",
    "message": "Shannon Coder call quota reached. Upgrade your plan at shannon-ai.com/plan"
  }
}

{
  "type": "error",
  "error": {
    "type": "rate_limit_error",
    "message": "Shannon Coder call quota reached. Upgrade your plan at shannon-ai.com/plan"
  }
}

Superfluxus contextus

Cum colloquium maius est quam fenestra contextus modelli, /v1/chat/completions et /v1/messages cum 400, genere invalid_request_error, codice context_length_exceeded et nuntio qui a prompt is too long incipit respondent. Instrumenta codicis hunc errorem legunt ut sessionem compactent et petitionem iterum mittant. A shannon-coder-1, shannon-1.6-lite et shannon-1.6-pro producitur.

{
  "error": {
    "type": "invalid_request_error",
    "code": "context_length_exceeded",
    "message": "prompt is too long: 140211 tokens > 131072 maximum",
    "param": null
  }
}

{
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "code": "context_length_exceeded",
    "message": "prompt is too long: 140211 tokens > 131072 maximum",
    "param": null
  }
}

Nuntius est prompt is too long: <sent> tokens > <limit> maximum. Cum duo numeri ignoti sunt, est prompt is too long. Please compact the conversation and try again.

In fluxu status iam est 200, itaque error ut eventus advenit et fluxus deinde modo normali clauditur:

data: {"error":{"message":"prompt is too long: 140211 tokens > 131072 maximum","type":"invalid_request_error","code":"context_length_exceeded","param":null}}

data: {"id":"chatcmpl-…","object":"chat.completion.chunk","created":…,"model":"shannon-coder-1","choices":[{"index":0,"delta":{},"finish_reason":"stop"}],"usage":{…}}

data: [DONE]

event: error
data: {"type":"error","error":{"type":"invalid_request_error","message":"prompt is too long: 140211 tokens > 131072 maximum"}}

event: message_delta
data: {"type":"message_delta","delta":{"stop_reason":"end_turn","stop_sequence":null},"usage":{…}}

event: message_stop
data: {"type":"message_stop"}

Quibus circulus agentis confidere possit

Solutio problematum

Clavis tua et saldum tuum in pagina Claves et usus sunt. Claves et usus