model/context: move context window config to host-level hive-c0re.nix
This commit is contained in:
parent
7e2f13cad8
commit
d3d52349c3
10 changed files with 81 additions and 59 deletions
|
|
@ -91,6 +91,7 @@ pub async fn approve(coord: Arc<Coordinator>, id: i64) -> Result<()> {
|
|||
¬es_dir,
|
||||
coord_bg.dashboard_port,
|
||||
&coord_bg.operator_pronouns,
|
||||
&coord_bg.context_window_tokens,
|
||||
)
|
||||
.await;
|
||||
drop(guard);
|
||||
|
|
@ -415,6 +416,7 @@ async fn sync_meta_after_lifecycle(coord: &Coordinator) -> Result<()> {
|
|||
&coord.hyperhive_flake,
|
||||
coord.dashboard_port,
|
||||
&coord.operator_pronouns,
|
||||
&coord.context_window_tokens,
|
||||
&agents,
|
||||
)
|
||||
.await
|
||||
|
|
|
|||
|
|
@ -73,6 +73,7 @@ pub async fn rebuild_agent(coord: &Arc<Coordinator>, name: &str, current_rev: &s
|
|||
¬es_dir,
|
||||
coord.dashboard_port,
|
||||
&coord.operator_pronouns,
|
||||
&coord.context_window_tokens,
|
||||
)
|
||||
.await;
|
||||
drop(guard);
|
||||
|
|
@ -160,6 +161,7 @@ pub async fn ensure_manager(coord: &Arc<Coordinator>) -> Result<()> {
|
|||
¬es_dir,
|
||||
coord.dashboard_port,
|
||||
&coord.operator_pronouns,
|
||||
&coord.context_window_tokens,
|
||||
)
|
||||
.await?;
|
||||
if let Some(rev) = current_rev {
|
||||
|
|
|
|||
|
|
@ -51,6 +51,13 @@ pub struct Coordinator {
|
|||
/// meta flake); the harness substitutes it into the agent /
|
||||
/// manager system prompt at boot.
|
||||
pub operator_pronouns: String,
|
||||
/// Per-model context-window sizes in tokens. Set via the host-level
|
||||
/// `services.hive-c0re.contextWindowTokens` NixOS option; injected
|
||||
/// into each container as `HIVE_CONTEXT_WINDOW_TOKENS_<KEY_UPPER>`
|
||||
/// by the meta flake renderer. The harness uses these to derive
|
||||
/// compaction / auto-reset watermarks and exposes the active value
|
||||
/// on `/api/state` as `context_window_tokens`.
|
||||
pub context_window_tokens: std::collections::HashMap<String, u64>,
|
||||
agents: Mutex<HashMap<String, AgentSocket>>,
|
||||
/// Agents whose lifecycle action (currently just spawn) is in flight.
|
||||
/// Read by the dashboard to render a spinner; cleared when the action
|
||||
|
|
@ -139,6 +146,7 @@ impl Coordinator {
|
|||
hyperhive_flake: String,
|
||||
dashboard_port: u16,
|
||||
operator_pronouns: String,
|
||||
context_window_tokens: std::collections::HashMap<String, u64>,
|
||||
) -> Result<Self> {
|
||||
let broker = Broker::open(db_path).context("open broker")?;
|
||||
let approvals = Approvals::open(db_path).context("open approvals")?;
|
||||
|
|
@ -152,6 +160,7 @@ impl Coordinator {
|
|||
hyperhive_flake,
|
||||
dashboard_port,
|
||||
operator_pronouns,
|
||||
context_window_tokens,
|
||||
agents: Mutex::new(HashMap::new()),
|
||||
transient: Mutex::new(HashMap::new()),
|
||||
dashboard_events,
|
||||
|
|
|
|||
|
|
@ -138,6 +138,7 @@ pub async fn spawn(
|
|||
notes_dir: &Path,
|
||||
dashboard_port: u16,
|
||||
operator_pronouns: &str,
|
||||
context_window_tokens: &std::collections::HashMap<String, u64>,
|
||||
) -> Result<()> {
|
||||
validate(name)?;
|
||||
if let Some(other) = port_collision(name).await {
|
||||
|
|
@ -154,7 +155,7 @@ pub async fn spawn(
|
|||
// before `nixos-container create` so the `--flake meta#<name>`
|
||||
// ref resolves.
|
||||
let agents = agents_after_spawn(name).await?;
|
||||
crate::meta::sync_agents(hyperhive_flake, dashboard_port, operator_pronouns, &agents).await?;
|
||||
crate::meta::sync_agents(hyperhive_flake, dashboard_port, operator_pronouns, context_window_tokens, &agents).await?;
|
||||
let container = container_name(name);
|
||||
let flake_ref = format!("{}#{name}", crate::meta::meta_dir().display());
|
||||
run(&["create", &container, "--flake", &flake_ref]).await?;
|
||||
|
|
@ -273,6 +274,7 @@ pub async fn rebuild(
|
|||
notes_dir: &Path,
|
||||
dashboard_port: u16,
|
||||
operator_pronouns: &str,
|
||||
context_window_tokens: &std::collections::HashMap<String, u64>,
|
||||
) -> Result<()> {
|
||||
// Sync the meta flake (idempotent — no-op when the rendered
|
||||
// flake matches disk) so a manual rebuild from the dashboard
|
||||
|
|
@ -280,7 +282,7 @@ pub async fn rebuild(
|
|||
// got added directly via `nixos-container create` outside
|
||||
// hive-c0re).
|
||||
let agents = agents_for_meta(None).await?;
|
||||
crate::meta::sync_agents(hyperhive_flake, dashboard_port, operator_pronouns, &agents).await?;
|
||||
crate::meta::sync_agents(hyperhive_flake, dashboard_port, operator_pronouns, context_window_tokens, &agents).await?;
|
||||
// Then bump just this agent's input — picks up whatever
|
||||
// `applied/<n>/main` currently points at (deployed/<latest>).
|
||||
// Commits the lock if it changed.
|
||||
|
|
|
|||
|
|
@ -62,6 +62,12 @@ enum Cmd {
|
|||
/// system prompt can mention them. Default: `she/her`.
|
||||
#[arg(long, default_value = "she/her")]
|
||||
operator_pronouns: String,
|
||||
/// Per-model context-window sizes, as JSON object mapping model-family
|
||||
/// short name to token count. Threaded into each container as
|
||||
/// `HIVE_CONTEXT_WINDOW_TOKENS_<KEY_UPPER>` env vars. Set via the
|
||||
/// `services.hive-c0re.contextWindowTokens` NixOS option.
|
||||
#[arg(long, default_value = r#"{"haiku":200000,"sonnet":1000000,"opus":1000000}"#)]
|
||||
context_window_tokens: String,
|
||||
},
|
||||
/// Spawn a new agent container directly (`hive-agent-<name>`). Bypasses
|
||||
/// the approval queue — use only as an operator on the host. For
|
||||
|
|
@ -109,12 +115,17 @@ async fn main() -> Result<()> {
|
|||
db,
|
||||
dashboard_port,
|
||||
operator_pronouns,
|
||||
context_window_tokens,
|
||||
} => {
|
||||
let cwt: std::collections::HashMap<String, u64> =
|
||||
serde_json::from_str(&context_window_tokens)
|
||||
.context("--context-window-tokens: invalid JSON")?;
|
||||
let coord = Arc::new(Coordinator::open(
|
||||
&db,
|
||||
hyperhive_flake,
|
||||
dashboard_port,
|
||||
operator_pronouns,
|
||||
cwt,
|
||||
)?);
|
||||
manager_server::start(coord.clone())?;
|
||||
// Idempotent pre-flight: rewrite pre-meta-layout applied
|
||||
|
|
|
|||
|
|
@ -66,13 +66,14 @@ pub async fn sync_agents(
|
|||
hyperhive_flake: &str,
|
||||
dashboard_port: u16,
|
||||
operator_pronouns: &str,
|
||||
context_window_tokens: &std::collections::HashMap<String, u64>,
|
||||
agents: &[AgentSpec],
|
||||
) -> Result<()> {
|
||||
let _guard = META_LOCK.lock().await;
|
||||
let dir = meta_dir();
|
||||
std::fs::create_dir_all(&dir).with_context(|| format!("create {}", dir.display()))?;
|
||||
|
||||
let new_flake = render_flake(hyperhive_flake, dashboard_port, operator_pronouns, agents);
|
||||
let new_flake = render_flake(hyperhive_flake, dashboard_port, operator_pronouns, context_window_tokens, agents);
|
||||
let flake_path = dir.join("flake.nix");
|
||||
let on_disk = std::fs::read_to_string(&flake_path).unwrap_or_default();
|
||||
let initial = !dir.join(".git").exists();
|
||||
|
|
@ -235,6 +236,7 @@ fn render_flake(
|
|||
hyperhive_flake: &str,
|
||||
dashboard_port: u16,
|
||||
operator_pronouns: &str,
|
||||
context_window_tokens: &std::collections::HashMap<String, u64>,
|
||||
agents: &[AgentSpec],
|
||||
) -> String {
|
||||
use std::fmt::Write as _;
|
||||
|
|
@ -283,8 +285,19 @@ fn render_flake(
|
|||
HIVE_PORT = toString port;
|
||||
HIVE_LABEL = name;
|
||||
HIVE_DASHBOARD_PORT = toString dashboardPort;
|
||||
HIVE_OPERATOR_PRONOUNS = operatorPronouns;
|
||||
HYPERHIVE_STATE_DIR = "/agents/${name}/state";
|
||||
HIVE_OPERATOR_PRONOUNS = operatorPronouns;"#,
|
||||
);
|
||||
// Per-model context-window env vars declared in the host-level
|
||||
// `services.hive-c0re.contextWindowTokens` option. Use a sorted
|
||||
// iterator for deterministic flake output (no spurious git diffs).
|
||||
let mut sorted_tokens: Vec<(&String, &u64)> = context_window_tokens.iter().collect();
|
||||
sorted_tokens.sort_by_key(|(k, _)| k.as_str());
|
||||
for (key, val) in &sorted_tokens {
|
||||
let upper_key = key.to_ascii_uppercase();
|
||||
let _ = writeln!(out, " HIVE_CONTEXT_WINDOW_TOKENS_{upper_key} = \"{val}\";");
|
||||
}
|
||||
out.push_str(
|
||||
r#" HYPERHIVE_STATE_DIR = "/agents/${name}/state";
|
||||
};
|
||||
}
|
||||
];
|
||||
|
|
|
|||
|
|
@ -83,6 +83,7 @@ pub async fn run(coord: &Arc<Coordinator>) -> Result<()> {
|
|||
&coord.hyperhive_flake,
|
||||
coord.dashboard_port,
|
||||
&coord.operator_pronouns,
|
||||
&coord.context_window_tokens,
|
||||
&agents,
|
||||
)
|
||||
.await
|
||||
|
|
|
|||
|
|
@ -77,6 +77,7 @@ async fn dispatch(req: &HostRequest, coord: Arc<Coordinator>) -> HostResponse {
|
|||
¬es_dir,
|
||||
coord.dashboard_port,
|
||||
&coord.operator_pronouns,
|
||||
&coord.context_window_tokens,
|
||||
)
|
||||
.await
|
||||
{
|
||||
|
|
@ -139,6 +140,7 @@ async fn dispatch(req: &HostRequest, coord: Arc<Coordinator>) -> HostResponse {
|
|||
¬es_dir,
|
||||
coord.dashboard_port,
|
||||
&coord.operator_pronouns,
|
||||
&coord.context_window_tokens,
|
||||
)
|
||||
.await;
|
||||
// Mirror auto_update::rebuild_agent — the manager wants
|
||||
|
|
|
|||
|
|
@ -57,6 +57,31 @@ in
|
|||
approval needed.
|
||||
'';
|
||||
};
|
||||
contextWindowTokens = lib.mkOption {
|
||||
type = lib.types.attrsOf lib.types.int;
|
||||
default = {
|
||||
haiku = 200000;
|
||||
sonnet = 1000000;
|
||||
opus = 1000000;
|
||||
};
|
||||
example = {
|
||||
haiku = 150000;
|
||||
sonnet = 900000;
|
||||
};
|
||||
description = ''
|
||||
Per-model context-window sizes in tokens. Each key is a
|
||||
model-family short name matched case-insensitively as a
|
||||
substring of the active model name at runtime (e.g. `"sonnet"`
|
||||
matches `"claude-sonnet-4-5"`). The defaults cover the known
|
||||
Anthropic families; add entries for new models or override
|
||||
existing ones here to change the window for all agents at once.
|
||||
|
||||
Passed to `hive-c0re serve` as JSON and injected into every
|
||||
container's harness service environment as
|
||||
`HIVE_CONTEXT_WINDOW_TOKENS_<KEY_UPPER>`. Changes propagate
|
||||
on the next `↻ R3BU1LD` — no per-agent approval needed.
|
||||
'';
|
||||
};
|
||||
};
|
||||
|
||||
config = lib.mkIf cfg.enable {
|
||||
|
|
@ -89,7 +114,7 @@ in
|
|||
];
|
||||
environment.HYPERHIVE_GIT = "${pkgs.git}/bin/git";
|
||||
serviceConfig = {
|
||||
ExecStart = "${cfg.package}/bin/hive-c0re --socket /run/hyperhive/host.sock serve --hyperhive-flake ${cfg.hyperhiveFlake} --dashboard-port ${toString cfg.dashboardPort} --operator-pronouns ${lib.escapeShellArg cfg.operatorPronouns}";
|
||||
ExecStart = "${cfg.package}/bin/hive-c0re --socket /run/hyperhive/host.sock serve --hyperhive-flake ${cfg.hyperhiveFlake} --dashboard-port ${toString cfg.dashboardPort} --operator-pronouns ${lib.escapeShellArg cfg.operatorPronouns} --context-window-tokens ${lib.escapeShellArg (builtins.toJSON cfg.contextWindowTokens)}";
|
||||
Restart = "on-failure";
|
||||
RestartSec = 2;
|
||||
RuntimeDirectory = "hyperhive";
|
||||
|
|
|
|||
|
|
@ -36,41 +36,6 @@
|
|||
'';
|
||||
};
|
||||
|
||||
options.hyperhive.contextWindowTokens = lib.mkOption {
|
||||
type = lib.types.attrsOf lib.types.int;
|
||||
# Canonical defaults for known Anthropic model families.
|
||||
# Override any entry in your agent.nix, or add new keys for
|
||||
# model families not listed here.
|
||||
default = {
|
||||
haiku = 200000;
|
||||
sonnet = 1000000;
|
||||
opus = 1000000;
|
||||
};
|
||||
example = {
|
||||
haiku = 150000;
|
||||
sonnet = 900000;
|
||||
};
|
||||
description = ''
|
||||
Per-model context-window sizes in tokens. Each key is a
|
||||
model-family short name (e.g. `"haiku"`, `"sonnet"`) matched as a
|
||||
case-insensitive substring of the active model name at runtime, so
|
||||
`"sonnet"` matches `"claude-sonnet-4-5"` and any future variant.
|
||||
|
||||
The defaults declared here cover the known Anthropic model families.
|
||||
Add or override entries in your `agent.nix` when using a
|
||||
non-standard model or when Anthropic changes a model's window.
|
||||
|
||||
Each entry is rendered as
|
||||
`HIVE_CONTEXT_WINDOW_TOKENS_<KEY_UPPER>` (e.g.
|
||||
`HIVE_CONTEXT_WINDOW_TOKENS_SONNET = "1000000"`). The harness
|
||||
checks these per-model vars in order (first substring match wins),
|
||||
then falls back to `200000` when no key matches. At runtime the
|
||||
effective window drives compaction (75%) and auto-reset (50%)
|
||||
watermarks, and is exposed via `/api/state` as
|
||||
`context_window_tokens`.
|
||||
'';
|
||||
};
|
||||
|
||||
options.hyperhive.allowedBashPatterns = lib.mkOption {
|
||||
type = lib.types.listOf lib.types.str;
|
||||
default = [ ];
|
||||
|
|
@ -264,25 +229,15 @@
|
|||
environment.etc."hyperhive/claude-plugins-auto-update.json".text =
|
||||
builtins.toJSON config.hyperhive.claudePluginsAutoUpdate;
|
||||
|
||||
# Model + context-window env vars consumed by the harness at boot.
|
||||
# HIVE_DEFAULT_MODEL seeds the initial model selection when no persisted
|
||||
# model choice exists in the state dir.
|
||||
# HIVE_CONTEXT_WINDOW_TOKENS_<KEY_UPPER> provides per-model overrides
|
||||
# (e.g. HIVE_CONTEXT_WINDOW_TOKENS_SONNET) from contextWindowTokens attrset.
|
||||
# SHELL must be set so claude's Bash tool finds a POSIX shell.
|
||||
environment.variables = lib.mkMerge (
|
||||
[
|
||||
{
|
||||
HIVE_DEFAULT_MODEL = config.hyperhive.model;
|
||||
SHELL = "${pkgs.bashInteractive}/bin/bash";
|
||||
}
|
||||
]
|
||||
++ lib.mapAttrsToList
|
||||
(model: tokens: {
|
||||
"HIVE_CONTEXT_WINDOW_TOKENS_${lib.toUpper model}" = toString tokens;
|
||||
})
|
||||
config.hyperhive.contextWindowTokens
|
||||
);
|
||||
# model choice exists in the state dir. SHELL must be set so claude's
|
||||
# Bash tool finds a POSIX shell.
|
||||
# HIVE_CONTEXT_WINDOW_TOKENS_* are injected by the meta flake from the
|
||||
# host-level `services.hive-c0re.contextWindowTokens` option — not set here.
|
||||
environment.variables = {
|
||||
HIVE_DEFAULT_MODEL = config.hyperhive.model;
|
||||
SHELL = "${pkgs.bashInteractive}/bin/bash";
|
||||
};
|
||||
|
||||
boot.isNspawnContainer = true;
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue