# Approvals + manager + helper events The approval queue is hyperhive's pivot: nothing that changes the shape of an agent (its config, whether it exists) happens without an operator click. The manager (`hm1nd`) is the policy gate in front of that queue; helper events are how it stays informed about what happens after a decision lands. ## End-to-end approval flow 1. Manager edits files under `/agents//config/` (any tracked path, but `agent.nix` is the contract entry point) and commits with its own git identity. 2. Manager submits the commit sha via `request_apply_commit(agent, commit_ref)`. `commit_ref` must be a commit **sha** (7-40 hex chars, short or full) — a branch or tag name is rejected so the approval pins an immutable commit. 3. **hive-c0re immediately fetches that commit from the proposed repo into the applied repo and tags it `proposal/`.** It resolves the sha locally against the proposed repo, fetches all of proposed's heads into applied's object db, then tags the resolved commit — `git fetch :` can't fetch by a bare sha (the left side of a refspec is a remote *ref name*), so the resolution happens on hive-c0re's side. The approval row stores both the manager-supplied sha and the canonical hive-c0re-vouched sha. From here on the proposed repo is irrelevant for this approval — the manager can amend, force-push, or `rm -rf` the proposed repo and the queued approval still points at an immutable git object inside applied. 3a. **Flake validation (ApplyCommit only):** after the proposal tag is planted, hive-c0re reads `proposal/:flake.lock` and runs two checks. If either check fails, no pending approval is created for the operator — the row is marked failed and surfaces on the dashboard with the validation message: - **Stale lock** — materialises the commit in a temp worktree, runs `nix flake lock` (no `--update-input` flags, so it only fills missing entries), and rejects if the committed `flake.lock` differs from the result. Triggered when the manager added or removed `inputs` in `flake.nix` without re-running `nix flake lock`. Fix: run `nix flake lock` in the config repo, commit, and re-submit. - **Duplicate inputs** — groups lock nodes by their canonical `original` field; rejects if two or more nodes share the same source. This usually means an input is missing `inputs..inputs.nixpkgs.follows = "nixpkgs"`. Fix: add the `follows` directive, re-lock, and re-submit. Both checks only flag *new* violations — agents whose lock already carried duplicates before this check was added are unaffected until a coordinated config-change pass via manager. 4. Operator sees the proposal as a card on the dashboard — a full multi-file diff, toggleable between three bases (vs the running tree / vs the last approved proposal / vs the previous queued proposal) — and clicks ◆ APPR0VE (or `hive-c0re approve ` on the CLI). 5. hive-c0re moves the working tree to `proposal/` and runs the build under a sequence of tags (see below). On success, `applied/main` fast-forwards to the proposal commit. On failure, main stays put and the working tree resets back to the previous deployed commit. 6. `HelperEvent::ApprovalResolved` (and `Rebuilt` for the ApplyCommit kind) land in the manager's inbox, carrying both the canonical sha and the terminal tag. ### Withdrawing a pending approval The manager can call `cancel_loose_end(kind: "approval", id)` to withdraw an approval that hasn't been acted on yet. The row transitions to `ApprovalStatus::Cancelled` (distinct from `Denied`/`Failed`), the dashboard pulls the card out of the pending pane, and `ApprovalResolved { status: "cancelled" }` fires on the manager + dashboard channels. Approvals that have already been approved/denied/failed return an error — the resolution is final once the operator (or a lifecycle failure) acted on the row. Sub-agent surface refuses the `approval` kind with a clear error: sub-agents don't submit approvals, so they have nothing of their own to withdraw. Manager-only. `InitConfig` approvals are the first step in a two-step spawn flow. On approve, hive-c0re seeds the proposed config repo with a default `agent.nix` template and sends the manager `HelperEvent::ConfigReady { agent }`. The manager then reviews, edits, and commits the template before calling `request_apply_commit` to proceed to an `ApplyCommit` approval. The first `ApplyCommit` creates the container; subsequent ones rebuild it with new config. This gives the manager (and operator) an explicit review gate on the initial configuration before any container is created. ### Approval kinds (wire shapes) `ApprovalKind` carries five variants; each maps to a different `commit_ref` encoding because that field is overloaded as the kind-specific payload carrier. - `ApplyCommit` — `commit_ref` is the manager-supplied git sha (7-40 hex chars). The canonical, hive-c0re-vouched sha after the proposal fetch lives in `fetched_sha` on the same `Approval` row (only `ApplyCommit` populates it). See the End-to-end flow above. - `Spawn` — direct container creation under the default `agent.nix` template. `commit_ref` is empty. Submitted via `HostRequest::RequestSpawn` (operator-gated, the `◆ R3QU3ST SP4WN` dashboard button + `hive-c0re request-spawn` CLI). The host-level `HostRequest::Spawn` variant bypasses the approval queue entirely — privileged-context use only (operator on the host shell, test scripts, one-off recoveries). The manager-side `RequestSpawn` is gone; managers go through the `InitConfig` → `ApplyCommit` two-step instead so the spawn captures their customised config. - `InitConfig` — `commit_ref` is empty; the variant just gates "seed the proposed repo with the default template" against operator approval. Step 1 of the two-step spawn flow above. - `UpdateMetaInputs` — `commit_ref` stores the JSON-encoded inputs array (`"[]"` = all inputs, `"[\"nixpkgs\"]"` = just nixpkgs, etc.). `agent` field is set to `hm1nd` (the requesting manager). On approve hive-c0re runs `nix flake update [inputs...]` on the meta flake and commits the resulting lock changes. - `SchedulePrompt` — `commit_ref` stores the JSON-encoded `SchedulePromptPayload` (target list, body, schedule) so the approval row carries the full submission verbatim. On approve hive-c0re inserts a row into `scheduled_prompts` with `source = approval:`; the worker fans the body out as inbox messages to each target at the scheduled time, recurring when `interval_seconds` is set. ### Scheduled prompts (submit paths) Two ways a row lands in `scheduled_prompts`: - **Operator-direct** (`source = "operator"`): the operator adds a schedule through the dashboard form. Lands in the table immediately, no approval gate — operator action is already the trust boundary. - **Agent-requested** (`source = "approval:"`): a sub-agent (or the manager) submits a `RequestSchedulePrompt` through the manager socket. An `ApprovalKind::SchedulePrompt` row is queued; on approve, hive-c0re inserts the schedule row with `source = approval:` so the audit trail points back at the operator decision (above). No self-target shortcut: even agent-self schedules need approval. The existing `remind` MCP tool stays the quick self-wake path (no approval, lands directly in the agent's own inbox); this module is the bigger, multi-recipient, operator-visible thing. ### Scheduled prompt worker (catch-up clamp) When hive-c0re comes back from being down, the worker sees rows whose `next_fire_at_unix` is well in the past. For recurring rows that would mean firing N delayed pulses in a row — spammy and useless. Instead the worker fires **once** per row and bumps `next_fire_at_unix` to the next interval slot ≥ `now`, recording how many cycles were skipped in `last_result` (per-target). Operators see "fired late, caught up from 17 skipped" instead of 17 wake-up storms. One-shot rows fire once (if past due, on the next worker pass) and are deleted by the worker; recurring rows survive until cancelled. `targets` is its own table (`scheduled_prompt_targets`) so partial cancellation flips a single row and the dashboard can show last-fired / last-result per recipient. Cancelling every target reaps the parent row on the next worker pass. ### Destroy semantics `HostRequest::Destroy { name, purge }` is the lifecycle tear-down, not an approval. Stops + removes the nspawn container, drops the systemd drop-in, fails any pending approvals. Persistent state (proposed/applied repos, claude credentials, `/state/` notes) is **kept by default** — recreating the agent with the same name reuses prior config + login. With `purge = true` the agent's `/var/lib/hyperhive/{agents,applied}//` trees are also wiped (config history + creds + notes gone forever). The manager refuses to destroy itself. ## Meta flake The hive-c0re-owned repo at `/var/lib/hyperhive/meta/` declares one flake input per agent (`agent-.url = "git+file:///var/lib/hyperhive/applied/"`) and one `nixosConfigurations.` output per agent. Each output wraps `inputs.agent-.nixosModules.default` with the identity + `HIVE_PORT` / `HIVE_LABEL` / `HIVE_DASHBOARD_PORT` injection module that `setup_applied` used to generate inline. Containers run against `--flake /var/lib/hyperhive/meta#`. Per-deploy lock flow (two-phase, owned by `actions::run_apply_commit` → `meta::{prepare,finalize,abort} _deploy`): 1. `meta::prepare_deploy(name)` runs `nix flake lock --update-input agent-` without committing. Working tree of meta now points the input at `applied//main` (which `run_apply_commit` already fast-forwarded to `proposal/`). 2. `lifecycle::rebuild_no_meta` runs `nixos-container update --flake meta#`. Nix evaluates against the staged lock. 3. On success — `meta::finalize_deploy(name, sha, "deployed/ ")` stages `flake.lock` and commits with `deploy deployed/ `. Meta's git log gains one entry per successful deploy. 4. On failure — `meta::abort_deploy()` runs `git restore flake.lock` so the meta history shows only successes; the failure stays as an annotated `failed/` tag in `applied/`. Single-phase variants exist for paths without rollback semantics: `meta::lock_update_for_rebuild(name)` for the manual `↻ R3BU1LD` button (commits if the lock changed) and `meta::lock_update_hyperhive()` for the auto-update flake-rev bump (one shot before per-agent rebuilds, commits if the lock changed). `meta::sync_agents(hyperhive_flake, dashboard_port, &agents)` is the idempotent reconciler called by `spawn`, `destroy`, `rebuild`, and the startup migration. Renders `flake.nix` from the agent list; if it differs from disk, runs `nix flake lock` + commits as `regenerate meta flake` (or `seed meta from N agent(s)` on the very first call). The manager has `/meta` RO-bound inside its container: `git -C /meta log --oneline` is the swarm-wide deploy log, `cat /meta/flake.lock | jq '.nodes["agent-"].locked'` resolves which sha each agent is pinned at right now. Dashboard surfaces the same info as a `deployed:` chip per container row. ## Two repos per agent ``` /var/lib/hyperhive/agents//config/ proposed — manager RW └── # any files the manager # wants in the commit. # agent.nix is the # convention entry # point; flake.nix is # tracked boilerplate # (manager doesn't edit # it). /var/lib/hyperhive/applied// applied — core-only ├── .git/ # tag-rich history ├── flake.nix # tracked, fixed │ # boilerplate exporting │ # nixosModules.default ├── agent.nix # working tree of main └── # also tracked /var/lib/hyperhive/meta/ swarm-wide flake — core ├── .git/ # one commit per successful │ # deploy ├── flake.nix # generated from agent set └── flake.lock # pins each agent's sha ``` Why two physical repos: the manager's `/agents//config/` is RW — a buggy or hostile agent can `git clean -fdx` its own proposed tree. The applied repo is never bind-mounted (except the read-only `.git` exposure described below) so a destructive move inside the container cannot reach it. The container's `--flake` ref is `/var/lib/hyperhive/meta#` (see "Meta flake" above). The agent's own `applied//flake.nix` is a fixed boilerplate that exports `nixosModules.default = import ./agent.nix`; the meta flake imports that module and wraps it with identity + `HIVE_PORT` / `HIVE_LABEL` / `HIVE_DASHBOARD_PORT`. ### Tag state machine Every approval id walks through a fixed set of tags on the underlying commit inside the applied repo: | Tag | When | Annotated? | |---|---|---| | `proposal/` | request_apply_commit, after fetch | no | | `approved/` | operator approve | no | | `building/` | rebuild started | no | | `deployed/` | rebuild succeeded — `main` ff's here | no | | `failed/` | rebuild failed | yes (body = error) | | `denied/` | operator deny | yes (body = operator note) | `applied/main` is always the latest `deployed/*`. `denied/` and `failed/` are terminal; the manager submits a new commit + new approval id to retry. Because tags are first-class git objects, rejected and failed trees stay browsable forever — `git log --tags` in the applied repo is the audit trail. ### Dispatch via `rebuild_queue` Long-running approval work — `ApplyCommit`, `UpdateMetaInputs`, `Spawn` — no longer runs inline inside `actions::approve`. Instead the approval handler enqueues a `QueueEntry` into the global `rebuild_queue`: | `ApprovalKind` | `QueueKind` queued | `QueueSource` | |---|---|---| | `ApplyCommit` | `Rebuild` | `Approval` | | `UpdateMetaInputs` | `MetaUpdate` | `Approval` | | `Spawn` | `Spawn` | `Approval` | | `InitConfig` | — runs inline (sub-second git seed) | — | | `SchedulePrompt` | — runs inline (single sqlite insert) | — | Each queue entry carries the originating `approval_id` so the worker can re-fetch the approval row when it dispatches, run the kind-specific pipeline (`run_approval_apply_commit` / `run_approval_update_meta_inputs` / `run_approval_spawn`), and fire the matching `HelperEvent::*` on completion via `finish_approval`. Two visible consequences: - **Operator dashboard**: after clicking APPR0VE the work-in-progress shows up on the *rebuild queue* card (`POST /api/state.rebuild_queue` + live `rebuild_queue_changed` events), not on the approvals panel (which already moved the row to "approved"). A long meta-update cascade renders as a parent entry with one child per per-agent rebuild — see `docs/web-ui.md` for the layout. - **Cancellation**: the dashboard's *× cancel* button on a `Queued` entry calls `POST /api/rebuild-queue/{id}/cancel`, which flips the entry to `Cancelled` before the worker dispatches it. Returns `{"cancelled": true}` on success, `{"cancelled": false}` if the entry already left `Queued` (running / done / failed) — terminal states can't be retroactively rewritten. `QueueSource::Approval` carries the `approval_id` so a tail-end build failure surfaces back as a failed approval row, not just a silent queue entry. `QueueSource::Manual` (dashboard ↻ R3BU1LD) and `QueueSource::AutoUpdate` (boot-time sweep) use the same queue but skip the approval row plumbing. ### Forge mirror When the bundled `hive-forge` container is running — on by default, `hyperhive.forge.enable` — hive-c0re mirrors every agent's applied repo into a private `agent-configs` Forgejo org. `forge::push_config()` pushes `applied/main` plus every tag to `agent-configs/` after each ref mutation: the spawn that seeds `deployed/0`, every `request_apply_commit` (which plants `proposal/`), every approve / deny, and a sweep at startup. Pushes are best-effort — a missing or stopped forge never blocks a deploy. The org is private and agents are not members, so only the `core` user (a Forgejo site admin) can read it: an agent can't reach another agent's config — or even its own — through the forge. The tokenised push URL is passed inline to `git push`, never written into `applied//.git/config`; that repo is RO-bind-mounted into the manager, and a stored token would leak core's admin credential to an agent. The dashboard deep-links into this org — a `config repo` link per container row and a `commit on forge` link per approval card. See `docs/web-ui.md`. ### Manager view of applied + meta The manager container gets three host-side bind mounts via `set_nspawn_flags`: - `/var/lib/hyperhive/agents/` → `/agents/` (RW) — proposed repos. Manager edits + commits per-agent config here. - `/var/lib/hyperhive/applied/` → `/applied/` (RO) — every agent's authoritative applied repo, including `.git`. - `/var/lib/hyperhive/meta/` → `/meta/` (RO) — the swarm-wide deploy flake. Each proposed repo (`/agents//config/`) is pre-configured with `applied` as a git remote pointing at `/applied//.git`. Useful incantations from inside the manager: ```sh git -C /agents//config fetch applied git -C /agents//config log applied/main --oneline git -C /agents//config show applied/refs/tags/deployed/ git -C /agents//config show applied/refs/tags/failed/ # body = build error git -C /agents//config show applied/refs/tags/denied/ # body = operator note git -C /agents//config rebase applied/main # base in-flight work on what's deployed git -C /meta log --oneline # swarm-wide deploy history cat /meta/flake.lock | jq '.nodes | with_entries(select(.key | startswith("agent-")))' ``` The RO binds block push at the kernel level, so the manager can only fetch / read — git plumbing inside the container cannot corrupt either authoritative repo. ## Migration from the pre-tag / pre-meta schemes Both overhauls (tag-driven flow + meta flake) ship in-place migrations that run on every hive-c0re startup. Idempotent; each phase is a no-op once already applied. Behaviour: - Tag-driven phase: assumes the operator ran the one-shot `git tag deployed/0 main` script (see commit history / earlier docs revisions) once per agent. Tagging is non-destructive: it doesn't touch live containers, state dirs, or claude creds. - Meta-flake phase: rewrites each `applied//flake.nix` to the module-only boilerplate, wires the `applied` remote in each proposed repo, bootstraps the meta repo from the current agent list, and `nixos-container update`s every container at `meta#`. The expensive last step is guarded by `/var/lib/hyperhive/.meta-migration-done` so it only runs once across hive-c0re restarts. Set `HIVE_SKIP_META_MIGRATION=1` on the service to defer. No state loss in either migration. claude creds, /state/ notes, the events DB, proposed history, and applied history all survive. The manager keeps its session; sub-agents stay logged in. ## Manager (`hm1nd`) is hive-c0re-managed The manager container runs through the **same lifecycle as sub-agents**. On `hive-c0re serve` startup, if `hm1nd` is missing, hive-c0re creates it. The manager's flake lives at `/var/lib/hyperhive/applied/hm1nd/`; its proposed config at `/var/lib/hyperhive/agents/hm1nd/config/`. Manager can edit its own `agent.nix` (visible inside the container at `/agents/hm1nd/config/`) and submit `request_apply_commit("hm1nd", )` for operator approval. Differences from sub-agents: - `flake.nix` extends `hyperhive.nixosConfigurations.manager` (vs `agent-base`). - Container name is `hm1nd` (no `h-` prefix). - Web UI port via `lifecycle::agent_web_port("hm1nd")` — same FNV-1a hash as every other agent (8100..8999 range). - `set_nspawn_flags` adds two extra binds: `/var/lib/hyperhive/agents` → `/agents` (RW) so the manager can edit per-agent proposed repos, and `/var/lib/hyperhive/applied` → `/applied` (RO) so the manager can `git fetch` deployed/failed/denied tags from any agent's authoritative applied repo (see "Manager view of applied" below). - First-deploy spawn bypasses the approval queue (manager is required infrastructure). - Per-agent socket lives at `/run/hyperhive/manager/`, owned by `manager_server::start`. **Migration note** (for older hosts): drop any `containers.hm1nd = { ... }` block from your host NixOS config. hyperhive creates and updates the manager itself. ## Manager policy From `hive-ag3nt/prompts/system.md` (`` block, rendered via `hive_ag3nt::prompt::render`): the manager does NOT rubber-stamp sub-agent config requests. It verifies (role match, package legitimacy, cheaper alternative, blast radius) before committing and calling `request_apply_commit`. For ambiguous cases or anything that needs human signal, the manager calls `ask(question, options?, multi?, ttl_seconds?, to?)` — queues the question and returns the id immediately. When `to` is omitted (or `"operator"`) the question shows up on the dashboard; when `to` is a sub-agent's name, the recipient receives a `HelperEvent::QuestionAsked` and answers via their own `answer` tool. Either way the answer arrives back as `HelperEvent::QuestionAnswered { id, question, answer, answerer }` in the asker's inbox. Storage is `hive-c0re::operator_questions` (sqlite) — same table, with a nullable `target` column (NULL = operator). Dispatch goes through `hive-c0re/src/questions.rs::{handle_ask, handle_answer}` so both the agent + manager surfaces stay aligned. The answer flow is: ``` POST /answer-question/{id} agent: Answer { id, answer } → OperatorQuestions::answer(_, _, "operator") → questions::handle_answer → notify_agent(asker, QuestionAnswered { → OperatorQuestions::answer(_, _, agent) answerer: "operator", ... }) → notify_agent(asker, QuestionAnswered { answerer: agent, ... }) ``` Two more paths resolve a pending question with a sentinel answer: - `POST /cancel-question/{id}` (✗ CANC3L button on the dashboard) resolves with `[cancelled]`. The manager sees a terminal state and can fall back. - `ttl_seconds` deadline: a tokio watchdog spawned at submit time fires `answer(id, "[expired]")` once the ttl runs out. Already- resolved races no-op. The dashboard surfaces a `⏳ MM:SS` chip on each pending question with a deadline. ## Helper events to the manager `Coordinator::notify_manager(&HelperEvent)` enqueues an inbox message from sender `system` with the event JSON in the body. The manager harness no longer short-circuits these — they drive a regular claude turn so the manager can react. Variants (`hive_sh4re::HelperEvent`): - `ApprovalResolved { id, agent, commit_ref, status, note }` — fired by `actions::approve` + `actions::deny` whenever an approval transitions to its terminal state. - `Spawned { agent, ok, note }` — `actions::approve` (first-time ApplyCommit-kind) + admin `HostRequest::Spawn` (deprecated). - `Rebuilt { agent, ok, note }` — `auto_update::rebuild_agent` (covers startup scan + manual `/rebuild` from dashboard) + `actions::approve` (ApplyCommit). - `Killed { agent }` — admin `HostRequest::Kill` + dashboard `/kill` + manager `Kill` MCP tool. - `Destroyed { agent }` — `actions::destroy`. - `ContainerCrash { agent, note }` — `crash_watch`: a previously- running container went away with no operator-initiated transient state (Stopping / Restarting / Destroying / Rebuilding) AND no such transient was cleared in the last 30s (`RECENT_TRANSIENT_GRACE` tombstone, three `POLL_INTERVAL`s — closes the race where a lifecycle op finishes between two crash-watch polls and the container shows briefly as "stopped without transient" before the next start). Manager can `start` it again or escalate. - `NeedsLogin { agent }` — sub-agent has no claude session yet. Manager can't act directly (interactive OAuth); typically flags the operator. - `LoggedIn { agent }` — sub-agent just completed login. Manager often greets the agent on this event. - `ConfigReady { agent }` — a new agent's proposed config repo was just seeded (post-`InitConfig` approval). The manager can now edit `/agents//config/agent.nix`, commit the changes, and submit `request_apply_commit` with the commit sha to create the container (first ApplyCommit also triggers spawn bookkeeping). - `NeedsUpdate { agent }` — sub-agent's recorded flake rev is stale. Manager calls `update(name)` to rebuild — idempotent, no approval required. - `QuestionAnswered { id, question, answer, answerer }` — dashboard `/answer-question/{id}` (answerer = `"operator"`), peer `Answer` request (answerer = agent name), or ttl watchdog expiry (answerer = `"ttl-watchdog"`, answer = `"[expired]"`). - `QuestionAsked { id, asker, question, options, multi }` — fired when an agent calls `Ask { to: Some(), ... }`. The recipient responds via `Answer { id, answer }` and the asker sees the matching `QuestionAnswered`. Optional `sha` field on `ApprovalResolved`, `Spawned`, and `Rebuilt` carries the canonical hive-c0re-vouched commit sha. Optional `tag` on `ApprovalResolved` and `Rebuilt` only — the spawn path always lands at `deployed/0`, so the tag is implicit and not echoed. The tag values for the variants that do carry it: `deployed/` / `failed/` / `denied/` for approval-driven flows; `approved/` for the rare bare-approval case where no underlying action runs. Both fields are `Option`: `None` on the rebuild paths that don't change the deployed commit (e.g. `auto_update::rebuild_agent` reapplying the existing main, or the dashboard `↻ R3BU1LD` button when the lock didn't move). When set, `git show ` against `/agents//applied.git` inside the manager container yields the exact tree that was referenced. To add a new event: new `HelperEvent` variant + call sites + update `prompts/system.md` (`` block, the lifecycle- event list) so the manager knows the new shape. ## Auto-update on startup `hive-c0re serve` runs `auto_update::run` in a background task right after opening the coordinator. It enumerates managed containers and rebuilds any whose recorded hyperhive rev differs from the current one — sub-agents and manager go through the same `lifecycle::rebuild` path. "Rev" = canonical filesystem path of `cfg.hyperhiveFlake`. Marker file: `/var/lib/hyperhive/applied/..hyperhive-rev`. If the flake input has no canonical path (e.g. a `github:` URL), auto-update is a no-op — rebuild manually. The dashboard surfaces pending updates per agent: a clickable "needs update ↻" badge appears whenever the marker differs from current rev. The badge POSTs `/rebuild/`, calling the same `auto_update::rebuild_agent` path so manual triggers and the startup scan can't drift. When at least one container is stale, a top-level `↻ UPD4TE 4LL` button appears that loops over every stale container.