Six places asserted the old design as fact, and none of them mention the
change by name -- the class of doc breakage that is found by asking what
a diff made untrue, not by grepping for a feature:
- swarmctl/README.md and swarm-authelia-bridge/README.md both described
their own private canonical store. The bridge's "known limitation"
section described the seam as unsolved; it is what this fixes, so it
becomes what both writers must uphold instead.
- docs/swarm/{sso,ui,secrets}.md described a rendered artifact.
- The repo CLAUDE.md entry for swarmctl said the same.
- docs/tools/swarmctl-cli.md is regenerated (CI diffs it against the
clap tree), picking up the removed --store flag.
Operator-facing where it is read: the hand-editing consequence (values
survive a rewrite, comments do not) is stated in sso.md, where an
operator is being told to edit the file, rather than only in a module doc.
Adds swarmctl markdown-docs (a hidden Verb, same clap-markdown +
hide=true shape as hivectl markdown-docs) and generates
docs/tools/swarmctl-cli.md from it. Wires a swarmctl-docs freshness
check into nix/checks.nix, same shape as hivectl-docs, diffing against
packages.swarmctl.
One real gotcha: PathArgs::resolve() reads required
SWARMCTL_AUTHELIA_* deployment env vars and errors if unset -
swarmctl markdown-docs must not go through that path (it needs none of
those vars, and the docs build runs it outside any real deployment).
Restructured main() so resolve() only runs for the User arm, not
unconditionally before the match.
Also links the new doc from docs/tools/README.md (new 'for the swarm
operator' section), CLAUDE.md's swarmctl bullet, and
docs/conventions.md's flake-check list.
Verified: cargo check/clippy -D warnings/test/fmt -p swarmctl all
clean; swarmctl markdown-docs diffs clean against the committed doc
(checked against both a plain cargo build and the actual nix build.
#swarmctl output); scripts/check-issue-refs.sh clean.
Project-bootstrap scope per the issue: static build as a nix pkg,
empty start page for now, functionality deferred until auth against
authelia is figured out. Stack (Preact + wouter + TypeScript + JSX)
matches the shell decision from the earlier framework-paths thread —
a real SPA shell with a router and deep links, distinct from the
per-hive dashboard's vanilla-JS + custom-element MPA.
- New npm workspace frontend/packages/swarm-ui: one route (/), a
wouter Switch/Route shell, a 404 fallback. Reuses @hive/shared's
colors.css/theme.css/base.css for visual consistency; no other
shared JS (the vanilla-JS el()/dom.js helpers are superseded by
Preact in this shell).
- nix/packages/swarm-ui.nix: its own buildNpmPackage derivation
(scoped to just this workspace via an explicit buildPhase), not
folded into nix/packages/frontend.nix's packages.default closure —
same reasoning swarm-controller/swarmctl already use for staying out
of daemonBins: a hive that doesn't run the swarm controller
shouldn't carry swarm-ui bytes.
- npmDepsHash recomputed in both frontend.nix and swarm-ui.nix (same
shared lockfile, new deps: preact, wouter-preact, typescript).
- Added swarm-ui to nix/checks.nix alongside frontend, for the same
FOD-staleness reason plus being the only thing that actually builds
it in CI (not in packages.default's closure like frontend is, so
nix flake check wouldn't otherwise touch it).
- npm run typecheck (tsc --noEmit) is available locally; not yet wired
into CI — esbuild transpiles TS without type-checking, so that's a
real gap, left as a follow-up rather than growing this bootstrap PR
with a new CI workflow step.
Verified: nix build .#swarm-ui and .#frontend both succeed; npm run
build (root, all workspaces) succeeds; tsc --noEmit clean;
scripts/check-issue-refs.sh clean.
The swarm-authelia module states that its users database is written by
swarm-controller, but nothing ever granted the means. This adds the tool
that does it.
swarmctl runs as root on the controller's host and acts directly. The
rootless alternative was examined and does not work: relocating the users
file into a directory the controller owns only turns a write problem into
a read problem, because authelia must then reach across the same boundary
in the other direction. Making that read work needs either a hand-pinned
gid or world-readable password hashes.
The user store is two files, one authoritative: users.json is canonical,
users.yml is a rendered artifact. That split is what lets the crate work
without a YAML parser -- the workspace has none, and adding one costs a
crates.io fetch, a lock update and a vendor hash for a schema we fully
control and only ever emit.
Passwords are generated by authelia rather than passed to it: argv is
world-readable, so a password on a command line is readable by any local
process for the lifetime of the call.
The three derived facts swarmctl needs about the authelia container --
machine, unit and the host-side users path -- become readOnly options on
the authelia module rather than literals repeated at the call site.
`docs/swarm.md` becomes `docs/swarm/README.md` and the shared-services
material moves to `docs/swarm/services.md`, following the shape
`docs/turn-loop/` and `docs/web-ui/` already use. The README keeps a
pointer so the reading path is unbroken.
Every referrer moved with it — five docs pages, two option descriptions
in swarm.nix, and CLAUDE.md's reading path. A pointer to a file that
moved is worse than one to a file that was deleted: the content still
exists, so the reader concludes the note is wrong rather than the path.
Per review: crate READMEs will be served on the docs page, so the same
technical content in docs/swarm.md and the repo map is redundancy, not
thoroughness.
docs/swarm.md keeps only what is operator-facing and specific to it --
the option, and why enable is not derived from services.hyperhive.enable
-- and points at the README for the socket-directory constraint. The repo
map keeps the one-line warning and the pointer, not the argument.
This is the same correction as the AgentWindow comments: I had written
the socket rationale into five places and called it coverage. Correcting
every copy is what preserves the cause.
The crate was a workspace member with no entry in CLAUDE.md, which is the
index that auto-loads into every turn -- a member missing from it is
invisible to everyone who comes after.
Both entries carry the socket-directory constraint rather than just
naming the daemon, because that is the one thing about this service a
reader can get wrong from a position that looks correct.
Prompted by hyperhive/hyperhive#3050: a reviewer held a formal
approve/request-changes verdict for CI to go green, which delayed a
merge that was only waiting on the review. CI, review approval, and
auto-merge arming are three independent gate conditions — a reviewer
is only responsible for the second one.
Documents the asymmetry that follows from auto-merge sometimes being
armed before a review lands: an unnecessary request-changes costs a
round-trip, but an approve given on lingering doubt can ship
immediately with no further human check. Default to holding when
unsure.
The crate described itself as a persistent scheduler and NodeId promised
stability across restarts. Neither is true: hive-c0re constructs an empty
Graph on every boot and re-derives desired state with its reconcile sweep,
and nothing in the workspace writes or loads a graph. hive-c0re's own
job_queue module doc has said "runtime-only (no persistence)" all along —
only the extracted library's prose drifted.
Seven claims corrected across the module doc, NodeId, the id-counter error
and the Graph type, plus the repo map. The module doc now states the fact
positively rather than just dropping the word: serde exists so the graph can
be projected onto a wire and so a store could be added later, ids and
timestamps are stable within a run.
NodeId spells out the consequence, since that is the part that could mislead
someone: an id stored outside the process is a historical record, not a
handle that will resolve after a restart.
Part of hyperhive#1898 (b): every docs subdir should have a top-level
README.md link, achieved by moving/renaming where an existing file
already fits the role.
docs/turn-loop.md already served as the hub + index for the three
sub-pages under turn-loop/ (claude-invocation.md, config.md, mcp.md),
so it moves wholesale rather than leaving a redundant top-level
pointer stub. Fixes every inbound/relative link across the repo
(top-level README.md, CLAUDE.md, docs/persistence.md,
docs/tools/scheduling.md, the sub-pages own back-link, hive-agent
README + doc comments, hive-agent/Cargo.toml, .prettierignore per-file
exemption entry) - grepped the whole tree for both turn-loop.md and
turn-loop/ to find every reference rather than trusting a partial
list.
nix fmt clean, cargo check -p hive-agent clean.
The wire types were in hive-host-sock, which is the host *socket* crate — so
anything living there is core-shaped by construction, and the projection had
quietly grown two core dependencies to match: it selected roots by matching
NodeKind::Dag, and rendered payloads through free functions in hive-c0re that
nothing obliged a second host to write.
hive-jobq is the wrong home too. That crate is the scheduler — logic — and
folding presentation in means every consumer of it carries a JSON vocabulary
it may never serve.
So: a new hive-jobq-wire. A host implements WireNode for its payload N and
WireResource for its resource name R; GraphWire::wire_snapshot is
blanket-implemented for Graph<N, R> when both hold, and for nothing else. A
payload that has never said how it displays has no way onto the wire.
wire_snapshot takes the roots to serve rather than reading Graph::roots
itself. Nothing is ever removed from a Graph, so retention is a policy only
the host can hold; hive-c0re passes visible_roots(), which is the existing
MAX_HISTORY_DAGS bound selected structurally (a root is a node with no
parent) instead of by node kind.
The operator's instruction on the issue was "the closure returns an array
of guids, and enqueue_job returns the node ids in that order". What was
here instead returned a HashMap of everything inserted, and no caller used
the keys: submit dropped the return, insert_group did into_values(), and
the scheduler ignored what append_subgraph handed back. The guid-keyed
lookup was dead weight, and into_values() made that Vec arbitrarily
ordered -- harmless only because nothing read it.
insert_job now takes FnOnce(&JobBuilder) -> Vec<NodeGuid> and returns the
matching ids positionally. A handle from another job is UnknownNode rather
than a silent omission: the return is positional, so a short vector would
misalign every id after it.
c0re's Declare stays FnOnce(&Job) and the wrapper names no handles in one
place, rather than ending seven templates in an empty vector -- a DAG is
addressed by its container node, which submit inserts itself. That frees
insert_group from needing every id, so the node_rt pre-seeding goes too:
NodeRuntime is one Option field and every reader already tolerated a
missing entry (entry().or_default(), get().and_then(), iter().find()).
The tests are the argument for the shape: capturing a handle through a
mutable binding to look it up in the map afterwards collapses into
returning it and destructuring the result.
This file auto-loads into every agent's context on every turn, so each
error in it is paid continuously by everyone rather than once by whoever
reads it. Four of them:
Five workspace members were missing entirely: hive-jobq, hive-types,
hive-agent-sock, hive-core-agent-sock, hive-screen-mcp. hive-jobq is the
worst of those to omit -- it is the crate the c0re-side job_queue layer
is being folded into, so the map that loads every turn did not mention
the target crate exists.
hive-claude was listed as a workspace member. It has no directory here
and is consumed as a dependency, so anyone following the map went
looking for source that is not in this repo. Moved to its own section
that says so.
The issue tracker was linked at a loopback address. Inside an agent
container loopback is the agent, not the forge, so the link cannot
resolve for its primary audience; point at the env var instead.
The ops label was written with a colon. The real label uses a slash,
and a colon does not fail loudly -- the filter matches nothing and
returns unfiltered results, which reads like a backlog instead of like
an error.
mara: the option was the wrong shape for the concern. "this host runs
nothing else" is a deployment expectation, not something a module
should assert about its own host — and asserting it made co-location
look like a config toggle rather than what it is.
Replaced with docs/snapshot-store.md, which the module had no docs
page at all before: enabling it, why the mesh is the authentication
(cryptokey routing already binds source address to pubkey, so certs
would authenticate the same fact twice and add an expiry), why the
destination is keyed per agent (a per-hive prefix splits an agent's
chain the first time it migrates), what the sender may and may not
choose, why the firewall rule is interface-scoped, what a snapshot
does and does not contain, and what the pull side still needs.
The dedicated-host expectation is stated there as an operational
assumption with its own failure mode — true on day one, quietly false
the day someone notices the box has spare disk — rather than as an
assertion someone flips to false to make the build proceed.
Linked from CLAUDE.md's reading paths.
hive-forge-notify grows a second binary, hive-github-notify. The two
share the notification half of the job — tolerant parse, classification,
formatting, dedupe, todo delivery — and nothing else: each binary owns
its host's protocol outright.
Two binaries rather than one multi-source daemon, and rather than a
cargo feature. A feature would unify across the workspace and cost every
crate its build cache. Two binaries keep the decision in nix: forge.nix
installs the forge unit, github.nix installs the github one under
hyperhive.github.enable, so a hive built without that module has no
github poller in its closure at all — GitHub access is separable (a
tier, a policy boundary), not merely switched off. Both binaries ship
from the existing derivation, so packages.nix is untouched.
The split is real at the code level too, not just at the unit level.
source.rs is a trait; the impls live in the binaries that use them, so
neither binary links the other's protocol code and the library names no
host at all. The forge-only assigned-issue rollup moves into the forge
binary for the same reason: it asks the forge what is assigned to this
agent, which is not a notification-protocol concern.
At runtime the github unit needs a PAT at <state>/github-token, the same
dashboard-provisioned token the gh wrapper and the git credential helper
already use. No PAT: it logs why and exits 0, which is why the unit is
Restart=on-failure and not always.
Forgejo's notifications API is modelled on GitHub's, so one tolerant
parse serves both — the differences (string thread ids, PullRequest vs
Pull) are absorbed by lenient deserializers rather than a second parse
path. Thread ids normalise to String at the parse boundary; they are
only ever opaque keys. Todo keys gain a per-source prefix so the two
hosts cannot collide, and the forge's is deliberately empty to keep
existing forge todo keys stable across the deploy that lands this.
The github loop honours the server's X-Poll-Interval, re-arming only
when the server asks for a slower cadence than ours; the hint is read
before the status check, because it arrives on error and empty pages too
and that is exactly when it matters. Reading the notification stream
needs the notifications scope on the PAT, which a token minted for push
access typically lacks; the failure mode is silence, so docs/github.md
says so explicitly.
Six places in the tree hand-rolled the same connect / write one JSON
line / read one JSON line back. Two of them — the harness serve loop's
client and the MCP server's — were byte-identical apart from a six-line
wrapper, ~145 lines of literal copy-paste. The other four each
reimplemented a subset, and the subsets had drifted: some named the
socket path in their errors and some did not, one classified transient
against fatal failures and the rest retried nothing at all, two drained
the response and two decoded it.
That duplication was defended when the daemons were split out, on the
grounds that a daemon's socket etiquette should stay visible in the
crate that depends on it. The etiquette genuinely does differ. The code
does not, and five copies is where "each daemon documents its own
etiquette" stops paying for itself.
`hive-sock-client` now owns the transport once, generic over the
request and response types so it is protocol-agnostic: the host-served
control socket and the harness's in-agent socket both use it with their
own wire-type crates. The two real differences become values instead of
forks. Retry is `Retry::RideOutRestart` (2/4/8/16/30s, sized to ride out
a service restart) for callers with no natural retry of their own, or
`Retry::None` for callers already inside a poll loop where the poll
interval is the retry — and the reason each caller picked one is a
comment at the call site rather than a reimplementation. The response is
either decoded (`request`) or half-closed and drained (`notify`, where
the drain exists so the server's write-back doesn't land on a closed
socket). Whether a failure propagates or is logged and swallowed stays
at the call site, because that is the caller's choice and not a property
of the transport.
Errors always name the socket path now, everywhere. That detail is
load-bearing: a permission problem on a socket that reads as "is the
daemon running?" sends the operator to fix the wrong thing.
The transient-against-fatal enum is gone rather than moved. Serialising
happens before the retry loop and deserialising after it, so only
connect, I/O and short-read failures can reach the loop at all — a
deterministic failure is now unretryable by construction instead of by
classification.
It is deliberately a new crate and not part of `hive-agent-sock`. The
`*-sock` crates are pure wire types by convention — `hive-agent-sock`
depends on serde and nothing else — and the two largest copies talk to
the host socket, whose types live in a different crate entirely. A
transport in either wire-type crate would drag tokio into it and point
the wrong way besides.
No wire-format change: same JSON line in, same line out.
The poller was a `tokio::spawn` inside the `hive-agent` serve loop. It
never needed anything from that loop except a socket path, so being
in-process bought nothing and cost two things: a harness restart took
forge notifications down with it, and the whole forge/HTTP dependency
tree was linked into the serve-loop binary.
It is now `hive-forge-notify`, a per-agent daemon with its own systemd
unit, a sibling of `hive-bash-daemon` and `hive-matrix-daemon`. Same
contract as those two: it reaches the harness only by upserting todos on
the in-agent socket, and nowhere else.
The module moves verbatim (`notify.rs`) — the formatters, the activation
gates, the dedupe map and all 33 tests are unchanged. Only the socket
call sites are rewritten, onto a small local `todo_client` rather than
the harness's. That mirrors what both sibling daemons already do, and
the etiquette differs on purpose: the harness's client carries a 60s
backoff schedule sized to ride out a hive-c0re restart, which its
callers need because they have no retry of their own. This poller's two
call sites both sit inside the 30s poll loop and both treat a failure as
"leave the thread unread, try next tick", so the poll interval already
is the retry; a second backoff would only stack sleeps and delay the
rest of the batch.
The unit is `Restart=on-failure`, not `always`. An agent with no forge
account is a supported configuration and the poller reports it by
logging why and exiting 0 — under `always` that clean exit would be a
restart loop on every forge-less agent.
`forgejo-api`, `url` and `time` drop out of `hive-agent`'s dependencies
with the module.
Also corrects docs that outlived the code they described: the persisted
`forge_cursor` field is long gone (forge's own read-state is the durable
record of what has been delivered), but `docs/persistence.md` and the
`harness_state` module docs still documented it as live.
add a clippy-discipline note to docs/conventions.md under 'Building &
local checks' and a short pointer bullet in CLAUDE.md so the rule is
visible at first read. covers the three most common patterns that surfaced
in practice (too_many_lines → extract helper, doc_markdown → backticks,
must_use → handle or discard) and gives a concrete worked example
(TurnAccum extraction in stats.rs).
README: add a "Overriding nixpkgs" section explaining hyperhive.inputs.
nixpkgs.follows and showing it in the quick-start flake example, since
hyperhive pins its own nixpkgs and consumers embedding it as a flake
input generally want to follow their host's nixpkgs instead.
CLAUDE.md: the repo map still described a hive-ag3nt/ directory
grouping hive-agent, hive-agent-mcp, and hive-agent-wake — that
directory doesn't exist; they're three separate top-level crates.
Also added the three wire-type crates split out of hive-sh4re
(hive-host-sock, hive-priv-sock) and hive-metric, none of which were
listed.
Wraps check-issue-refs.sh and check-comment-blocks.sh so a push that
would fail the tracker-tag or comment-block CI jobs is caught locally
before it reaches the remote — eliminating the fix-and-repush round-trip.
Install once per clone:
ln -sf ../../scripts/pre-push .git/hooks/pre-push
Also adds a one-liner hint to CLAUDE.md so it surfaces in every
agent's context at session start.
Add docs/observability.md covering all services.hyperhive.otel.*
options: enable, endpoint, protocol, headersCredential,
extraResourceAttributes, debug (new in cb0a66147a), and
metricIntervalMs.
Includes:
- Built-in OTEL_RESOURCE_ATTRIBUTES labels (service.name, agent, hive, swarm)
- Cumulative temporality note (avoids Prometheus DELTA drop)
- Network note for host-side collectors on non-standard ports,
cross-referencing docs/network.md exposeHostPorts
Also:
- CLAUDE.md: add reading-path entry for the new doc
- docs/network.md: link the OTEL mention to observability.md
Closes no issue — gap found during doc sweep.
Add an 'Agent trust model' framing section at the top of docs/security.md
covering the conceptual model the existing mechanism-level sections serve:
- trust boundary = the container, not credential storage (agents have
passwordless sudo by default; isolating creds from the agent itself is
not a goal — cross-tenant leakage is)
- scoped per-agent forge/matrix tokens bound the blast radius
- threat model: prompt injection -> confused deputy (untrusted input is
the adversary; the agent + its tools are trusted)
- branch protection (agents push, operator merges) as the human-in-loop
checkpoint, incl. external VCS
- capability = accepted risk ('don't grant what you can't afford to lose')
- no auto-sandboxing of external tokens (operator-accepted scope)
Emerged from the 2026-06-24 security discussion. Updates the CLAUDE.md
pointer for findability.
Per #1407 review: the options reference should be markdown here; the
HTML + CSS belongs in the website repo where the theme lives and the
stylesheet can be shared with /docs.
- nix/docs/default.nix: drop the cmark-gfm HTML rendering, the inline
<style>, and the HTML index. Emit index.md + host.md + agent.md only
(CommonMark from nixosOptionsDoc).
- Remove nix/docs/style.css.
- Update docs/gotchas.md + CLAUDE.md to describe the md-only output and
point at the website for HTML rendering.
The website (hyperhive/website#26) renders these .md to themed HTML for
/options/, sharing one docs.css with the prose /docs/ tree.
The CSS theme-variable reference documents the dashboard/agent web-UI
stylesheets, so it belongs under docs/web-ui/ alongside shape/dashboard/
agent. Wire it into the web-ui.md index (split list + reading path) and
the CLAUDE.md file map. Closes#1405.
The internal/knowledge forge repo and its /knowledge container bind-mount
have been live since feat(#1167) but had no docs/ entry — only a mention
in the agent system prompt.
Add docs/knowledge.md covering:
- Agent access (/knowledge read-only bind-mount inside every container)
- Repository layout (internal/knowledge, auto-seeded README)
- Sync mechanism (webhook on push + periodic pull fallback)
- State paths (/var/lib/hyperhive/knowledge on host)
- Contribution workflow (fork + PR, no direct push)
Also add a file-map entry and reading-path bullet in CLAUDE.md.
hivectl has verbs scattered across CLAUDE.md and domain docs (gateway.md
has the gateway subcommands, matrix.md has a brief mention) but no
single cheatsheet. Add docs/tools/hivectl.md covering all subcommands:
- forge create-user: provision/refresh agent or human forge accounts
- matrix create-user / sync-admin / promote-user / reset-password
- gateway create-user / delete-user / list-users (htpasswd management)
- agents restart / restart-all (container lifecycle via admin socket)
- choom <name> [--fresh]: interactive claude session in agent container
Also add a reading-path entry to CLAUDE.md pointing at the new file.
