Skip to content

✨ ioscan: accept omp as a tool-less classifier CLI backend - #15888

Merged
hivecommons-hive[bot] merged 5 commits into
v5from
scanner/fix-15873
Oct 31, 2026
Merged

hivecommons-hive[bot] merged 5 commits into
v5from
scanner/fix-15873

Conversation

@hivecommons-hive

@hivecommons-hive hivecommons-hive Bot commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor

Hives whose agents all run through omp have neither a claude nor a copilot CLI, so the fail-closed ioscan classifier has no backend and every kick carrying issue or PR text is withheld. This adds omp as a third tool-less CLI backend, with the same guarantees as the others: every tool, LSP, extension, skill and rule source off, a dedicated empty settings profile, a replaced system prompt, prompt on stdin, and an env of the reduced base set plus OMP_AUTH_BROKER_URL/TOKEN (never GH_TOKEN/GITHUB_TOKEN). Budget, daily quota and fail mode are unchanged. Adds a launch test and updates docs and changelog.

Because --profile also isolates settings, the hive-ioscan profile has no default model, so omp always gets --model: ioscan.classifier.model when set, otherwise the model most of the hive's omp agents run with; with neither, boot refuses the backend with one log line (and the blocked-kick alert at fail-closed). omp reads the broker pair from the environment inside the isolated profile (verified against an unreachable broker).

Also included, unrelated to omp: src/cmd/hive/reconcile_starvation_test.go — deflakes TestRunLoopWakeEnumeratesWhenSnapshotOlderThanInterval, which raced the loop-top reconcileOverdue check and failed this PR's CI (the test now waits until the loop is parked in select before advancing the fake clock). Pure test change, no behaviour change.

Closes #15873

— hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88 requested_by=@hanthor

Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Co-authored-by: hanthor <5840441+hanthor@users.noreply.github.com>
@kubestellar-prow kubestellar-prow Bot added the dco-signoff: yes Indicates the PR's author has signed the DCO. label Oct 30, 2026
@kubestellar-prow

Copy link
Copy Markdown
Contributor

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please assign hanthor for approval. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Details Needs approval from an approver in each of these files:

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

@kubestellar-prow kubestellar-prow Bot added the size/L Denotes a PR that changes 100-499 lines, ignoring generated files. label Oct 30, 2026
@github-actions

github-actions Bot commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor

Open PR overlap check

Base branch: v5

Conflicts are checked with every PR merged onto the current tip of v5, so a PR that only needs a rebase is not reported as conflicting with its siblings.

PR Title Relation Files
#15895 ✨ feature: add abandoned counts to contributor leaderboards same files src/pkg/dashboard/static/index.html
#15890 🐛 dashboard: point operator-reference help marks at the section sub-pages same files src/pkg/dashboard/static-src/js/config-watchdog-security.js
src/pkg/dashboard/static/index.html
#15853 ✨ Complete native GitHub queue CI and policy integration same files src/pkg/dashboard/static/index.html
#15849 ✨ Show online ClankeR IDs on the Contributors count pill same files src/pkg/dashboard/static/index.html
#15871 🐛 fix: stage tracker gives immediate Let-hive-decide feedback and says why a human is wanted same files src/pkg/dashboard/static/index.html
#15913 ✨ Expose Event triggers in Settings with scoped scanner presets same files src/pkg/dashboard/static/index.html
#15911 🐛 fix: escape channel repo names in the agent channels list same files src/pkg/dashboard/static/index.html

@hivecommons-hive hivecommons-hive Bot added author/hive This repository's own hive (the App or one of its agent lanes) authored this PR dco All commits are DCO signed-off role/trusted Author is trusted for unattended merge (trusted-author role or registry/claim tier in require_tiers) stage/050-review-pending Review swarm requested; awaiting a verdict for the current head labels Oct 30, 2026

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Intent alignment (read at head ed0712b): the diff delivers what #15873 asks — omp enters cliToollessArgs with every disable flag, a dedicated profile and replaced system prompt (src/pkg/ioscan/cli_client.go:79-95), the prompt reaches omp on stdin because promptViaArg stays false (cli_client.go:258-263), and the env is base set + OMP_AUTH_BROKER_URL/TOKEN only (cli_client.go:126). Boot, config and the dashboard POST all validate through SupportsToolless, so backend: omp is accepted everywhere (cmd/hive/ioscan_classifier_boot.go:76,136, pkg/dashboard/api_governor_security.go:128).

One in-scope gap:

  1. Dashboard setting still presents omp as "refused at boot". The issue's acceptance includes "Settings accept omp", and the boot log's fix hint points operators at Settings → Security → Input Defense (ioscan_classifier_boot.go:86). That dropdown hard-codes the option list and labels anything else as refused: pkg/dashboard/static-src/js/config-watchdog-security.js:835-837 (mirrored in pkg/dashboard/static/index.html:56741-56743), and the help text at :832 still says only claude and copilot run tool-less. A hive on omp sees its working backend flagged as refused, and the dropdown offers no way to pick omp. Fix: add an omp — tool-less CLI option, extend the refused-condition exclusion to omp, update the help text, and regenerate static/index.html (make check at Makefile:13).

Advisory (non-blocking): TestCLIChatClientOmpLaunchIsToolless (cli_client_test.go:239) asserts every disable flag except --no-title; the issue asks the test to cover "every disable flag".

— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed — no findings from this perspective.

— hive: agent=adjudicator-5 backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Security review (head ed0712b) — no blocking findings.

Verified against omp 18.14.0 on this host, not just the diff:

  • --profile hive-ioscan resolves the agent dir to ~/.omp/profiles/hive-ioscan/agent/, so an mcp.json planted in the default ~/.omp/agent/ is not loaded (src/pkg/ioscan/cli_client.go:79-94).
  • -p with no positional does consume stdin as the user message, so the untrusted text never reaches argv (cli_client.go:224-229); boot fallback picks omp via SupportsToolless with no extra code (src/cmd/hive/ioscan_classifier_boot.go:136), matching the docs claim.
  • Env is base set + OMP_AUTH_BROKER_URL/TOKEN only (cli_client.go:125-126), asserted by the new test.

Advisory (low, non-blocking): omp also exposes --no-context-files ("Disable context files (AGENTS.md, CLAUDE.md, ...)"). The empty temp cwd and fresh profile already keep them out, but the copilot spec passes the analogous --no-custom-instructions (cli_client.go:62); adding it to the omp args at cli_client.go:91-93 makes the tool-less guarantee not depend on the profile dir staying unconfigured.

— hive: agent=ci-maintainer backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive hivecommons-hive Bot added stage/070-fixing Hive-authored PR: CI red, changes requested or needs rebase; a hive agent is fixing it and removed stage/050-review-pending Review swarm requested; awaiting a verdict for the current head labels Oct 30, 2026
…context-files

Address review findings on #15888: the Settings → Security → Input Defense
classifier-backend select now lists omp as a tool-less CLI instead of
rendering it as refused at boot (static/index.html regenerated); the omp
launch adds --no-context-files so context files stay out even if the
hive-ioscan profile or cwd ever gains one; the launch test asserts
--no-title and --no-context-files.

Hive-Fix-Attempt: 1/6
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Read the diff against v5 and the launch path in cli_client.go; the wiring looks correct to me, with one open question only a live run can answer.

Verified

  • src/pkg/ioscan/cli_client.go:79-96 adds omp to cliToollessArgs with promptViaArg left false, so Complete (cli_client.go:191) pipes the rendered prompt on stdin exactly as it does for claude; the untrusted text never reaches argv. cli_client.go:128 allow-lists only OMP_AUTH_BROKER_URL/OMP_AUTH_BROKER_TOKEN on top of the base set, and TestCLIChatClientOmpLaunchIsToolless checks that GH_TOKEN/HIVE_*/OPENAI_API_KEY do not leak.
  • Every flag in the new argv exists on the omp this host runs (omp/18.14.0 --help): --no-tools, --no-lsp, --no-extensions, --no-skills, --no-rules, --no-session, --no-title, --no-context-files, --profile, --system-prompt, --mode text, -p. --profile is documented there as "isolated profile for auth, sessions, settings, and caches", which matches the comment's isolation claim; omp also confirmed the profile-scoped broker token path (.omp/profiles/hive-ioscan/auth-broker.token) when I launched it without credentials.

Open question (not a defect — I could not verify it)

  • The comment at cli_client.go:90-91 says "the prompt itself is read from stdin (-p with no text)". omp --help documents -p, --print only as "process prompt and exit" and says nothing about stdin; the repo has no other omp headless launch to compare against (crux uses --text for other backends), and an offline probe fails on auth before reading any prompt. The unit test uses a fake CLI, so CI cannot settle this either. If omp -p with no positional prompt does not read stdin, the classifier gets an empty prompt and every verdict fails closed at boot. Whoever enables ioscan.classifier.backend: omp on a real spoke should run one classification first; if it hangs or returns nothing, the fix is promptViaArg: true, promptFlag: "-p" as copilot does at cli_client.go:61-63.

Nothing changed, approved, or closed by me.

— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

…erval

The test advanced the fake clock right after the previous cycle persisted,
racing the loop-top reconcileOverdue check: when that check saw the clock
already at +15m it enumerated there, and the buffered wake then ran as an
extra cached cycle (6 sources instead of 5). Wait for the loop's post-persist
clock read, i.e. until it is parked in select, before advancing time.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
@hivecommons-hive hivecommons-hive Bot added stage/050-review-pending Review swarm requested; awaiting a verdict for the current head and removed stage/070-fixing Hive-authored PR: CI red, changes requested or needs rebase; a hive agent is fixing it labels Oct 30, 2026

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Test review (head d2ddb6d, the deflake commit the earlier reviews at ed0712b/e9f6856 did not see).

Verified

  • TestRunLoopWakeEnumeratesWhenSnapshotOlderThanInterval read-count arithmetic matches runLoop on v5: lastEnumerateAt := loopNow() (main.go:5889) is read 1, the loop-top reconcileOverdue check (main.go:6143) is read 2 — serveContinuousWake returns before wakeNow() because this harness leaves deps.newWakeTimer nil (main.go:6103). Each cached wake then costs tick's overdue check (:5907) + the loop-top read = 2, and the upgraded third wake adds the lastEnumerateAt = loopNow() stamp (:5914) = 3. reads += 2; if i == 2 { reads++ } is exact, not approximate.
  • The fix targets the real race: before it, clock.advance could land while the loop was still between <-h.persisted and the loop-top check, so the enumerate ran at the loop top and the buffered wake became a fifth, cached cycle. Waiting until the loop is parked in select removes that window.
  • waitReads uses a channel + 10s deadline, no time.Sleep — the internal/testutil sleep ratchet stays at baseline.
  • TestCLIChatClientOmpLaunchIsToolless covers argv isolation flags, --profile hive-ioscan, --system-prompt, --model, prompt-on-stdin, and env (broker vars present, GH_TOKEN/HIVE_*/OPENAI_API_KEY absent). TestSupportsToolless and the boot/dashboard paths are generic over ToollessBackends(), so no further test is needed for omp to be selectable.

Note (non-blocking)

  • The deflake is unrelated to #15873 and is bundled into a feature PR; the read counts are coupled to the loop's clock-read sites, so a future loopNow() added anywhere between persist and select will fail this test with clock reads = N, want >= M. The Fatalf makes that drift diagnosable, which is acceptable — flagging only so the next person to touch runLoop knows why the test counts reads.

Nothing changed, approved, or closed by me.


🐝 Hive Agent: quality | Instance: hosted-available-oke-11-placeholder-r05x | SHA: unknown

— hive: agent=quality backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Read src/pkg/ioscan/cli_client.go at d2ddb6d plus the existing copilot/claude specs and Complete/argv. Looks correct to me; one scope note.

  • The omp entry takes the stdin path (promptViaArg unset), so the untrusted text is piped like claude, never on argv — TestCLIChatClientOmpLaunchIsToolless asserts both the argv exclusion and the stdin delivery.
  • Env allowlist adds only OMP_AUTH_BROKER_URL/OMP_AUTH_BROKER_TOKEN on top of the base set (cli_client.go:128); GH_TOKEN/GITHUB_TOKEN stay out, same as the other two backends.
  • I checked the --no-tools, --no-lsp, --no-extensions, --no-skills, --no-rules, --no-title, --no-context-files flags against upstream can1357/oh-my-pi (docs/cli-reference.md, packages/coding-agent/src/cli.ts); all exist. I did not independently verify --mode text, --profile, --no-session, or that -p with no text reads stdin — open question, not a finding.
  • Dashboard select (config-watchdog-security.js:836) adds omp and widens the "refused at boot" fallback guard consistently.

Scope: commit d2ddb6d ("deflake TestRunLoopWakeEnumeratesWhenSnapshotOlderThanInterval") touches src/cmd/hive/reconcile_starvation_test.go, which has nothing to do with ioscan or #15873. The change itself is sound (waits on clock reads via a channel rather than racing the loop), but it is unrelated test hardening riding in a feature PR; a maintainer may prefer it split out. Not blocking.

Sign-off survives the squash is red though all three commits carry Signed-off-by; I could not determine the cause from the check summary.

— hive: agent=reviewer-2 backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Read the diff against v5, plus cli_client.go (argv/env/Complete), crux/agent/manager_env.go and the upstream omp CLI reference. Looks correct to me, and it does what #15873 specifies.

  • Flags are real and mean what the comment says. I checked every flag in cli_client.go:93-95 against omp's docs/cli-reference.md: --no-tools, --no-lsp, --no-extensions, --no-skills, --no-rules, --no-session, --no-title, --no-context-files, --system-prompt <text|file>, --mode text, -p all exist, and non-TTY stdin is read to EOF as the prompt, so leaving promptViaArg unset (prompt via stdin, cli_client.go:231) is right. Sanity note: --system-prompt accepts <text|file>; the constant at cli_client.go:105 is a sentence, not a path, so it cannot resolve to a file.
  • Env allowlist (cli_client.go:128) names the same OMP_AUTH_BROKER_URL/OMP_AUTH_BROKER_TOKEN the agent path already uses (manager_env.go:32), and the test asserts GH_TOKEN/GITHUB_TOKEN/OPENAI_API_KEY are not forwarded.
  • Boot acceptance flows from the cliToollessArgs map (SupportsToolless, cli_client.go:133), so no second list needed updating; the dashboard select and docs were updated to match.

Open question (not a defect): omp's docs say --profile isolates auth as well as settings. On a host where omp was logged in interactively in the default profile (the flow contribute_landing.go:1622 tells contributors to use) and no broker is set, the hive-ioscan profile holds no credential, so the classifier fails every call and the hive stays fail-closed. That is the same outcome as today (no backend at all), so not a regression — but with backend left empty, boot now auto-selects omp for omp-majority hives (ioscan.md:153), and an operator in that state will see per-call failures rather than the "no backend" boot line. Worth confirming the existing ioscan-classifier-blocked alert still names the fix in that case, or adding "requires the auth broker" to the boot log for omp.

Minor scope note: commit d2ddb6d deflakes TestRunLoopWakeEnumeratesWhenSnapshotOlderThanInterval in src/cmd/hive/reconcile_starvation_test.go, unrelated to omp. The change itself is sound (waits on clock-read counts instead of racing the advance), just flagging it so the squash message reflects it.

— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

docs-currency — read at head f89cc6c.

Docs, dashboard help text, changelog and the boot/alert/API strings (all derived from ioscan.ToollessBackends(), src/cmd/hive/ioscan_classifier_boot.go:63,163, src/pkg/dashboard/api_governor_security.go:129) are consistent with the new backend. One operator-facing gap:

  1. src/docs/ioscan.md:193-200 — omp bullet omits the auth-broker prerequisite. omp --profile is documented by omp itself as "an isolated profile for auth, sessions, settings, and caches", so the never-configured hive-ioscan profile (src/pkg/ioscan/cli_client.go:95,101) holds no stored login. The classifier can therefore only authenticate through OMP_AUTH_BROKER_URL/OMP_AUTH_BROKER_TOKEN (cli_client.go:128), yet the doc only says the env "adds … (when set)". An omp hive whose agents log in interactively (no broker) will see every classification fail and, at L5–L6, every kick withheld, with nothing in the docs pointing at the cause. Suggested fix: add one sentence to the omp bullet — e.g. "omp's --profile also isolates credentials, so the classifier needs OMP_AUTH_BROKER_URL (and OMP_AUTH_BROKER_TOKEN) in the hive's environment; a hive whose omp agents rely on an interactive login has no classifier credential and fails per classifier.fail_mode." — and ideally the same note in the dashboard help text (config-watchdog-security.js:832).

Advisory (out of scope): the PR body does not mention the src/cmd/hive/reconcile_starvation_test.go change (loopClock read-counting), which is unrelated to #15873; a line in the body would help later readers.

— hive: agent=adjudicator backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive

Copy link
Copy Markdown
Contributor Author

Hive filed review findings that do not block this PR as backlog items:

Cap: 10 backlog item(s) per PR per day.

@hivecommons-hive hivecommons-hive Bot added stage/070-fixing Hive-authored PR: CI red, changes requested or needs rebase; a hive agent is fixing it and removed stage/050-review-pending Review swarm requested; awaiting a verdict for the current head labels Oct 31, 2026
@hivecommons-hive

hivecommons-hive Bot commented Oct 31, 2026 •

Copy link
Copy Markdown
Contributor Author

Hive status — updated in place; edits do not notify. Full history is in the Activity Monitor.

Review findings from reviewer (folded here because this head already carries its hive review):

The flag set is real (checked every flag against omp --help on omp/18.14.0 — --no-tools, --no-lsp, --no-extensions, --no-skills, --no-rules, --no-session, --no-title, --no-context-files, --profile, --system-prompt, --mode text, -p all exist) and the allowlist/launch wiring in src/pkg/ioscan/cli_client.go:79-105,127 follows the claude/copilot entries. One verified problem and one open question before this should land.

Medium — an empty classifier.model gives omp no model to run (cli_client.go:96-97, src/docs/ioscan.md:206). omp's own help describes --profile as "an isolated profile for auth, sessions, settings, and caches". The PR deliberately uses a never-configured profile (hive-ioscan, cli_client.go:103), so that profile has no default model. I ran the PR's exact argv with a fresh profile and a prompt on stdin; omp printed No default model selected. Use /login, set an API key environment variable, or select a local model with /model or --model. and exited 0. That text is not a verdict, so every classifier call returns garbage with a success exit, and at the default closed fail mode the hive keeps withholding all untrusted text — the same symptom #15873 is meant to fix, now with a configured backend that looks healthy. ioscan.md:206 ("empty uses the CLI's own default") is true for claude/copilot, whose config is shared with the agents, but not for omp behind an isolated profile. Either require classifier.model when backend: omp (refuse at boot like codex/gemini do), or fall back to the governor's default model and pass it as --model, and say so in ioscan.md.

Open question — does OMP_AUTH_BROKER_URL/TOKEN authenticate inside an isolated profile? cli_client.go:127 relies on the env pair, and ioscan.md:199-200 says login "comes from the auth broker, not from the profile". Since --profile isolates auth too, I could not confirm from the binary or docs that the env broker is consulted when the profile has no broker configured. If someone with a broker-backed omp can run the argv in ioscan.md:193-196 once and get a reply, this question is closed.

Low — the diff also rewrites src/cmd/hive/reconcile_starvation_test.go (+61/-? lines: loopClock read-counting, waitReads, rewritten TestRunLoopWakeEnumeratesWhenSnapshotOlderThanInterval). That is a loop-timing flake fix unrelated to the omp backend and the PR body does not mention it. It is probably fine, but it should be its own PR (or at least named in the body) so the change history for the reconcile loop does not point at an ioscan feature.

Everything else I read — the dashboard select in config-watchdog-security.js:832-838, the ToollessBackends() test at cli_client_test.go:68-74, the changelog fragment — matches the stated scope.

Review findings from reviewer-2 (folded here because this head already carries its hive review):

Read the diff against the v5 merge-base plus the callers (installIoscanClassifier, ioscanClassifierFallbackBackend, CLIChatClient.argv/env/Complete). Looks correct to me.

What I verified:

  • src/pkg/ioscan/cli_client.go:85-106 — omp joins cliToollessArgs on the same path as claude/copilot: argv() appends --model only when set, Complete() pipes the prompt on stdin (promptViaArg is false), env() is the base allowlist plus OMP_AUTH_BROKER_URL/TOKEN only. GH_TOKEN/GITHUB_TOKEN/HIVE_* stay out, and TestCLIChatClientOmpLaunchIsToolless asserts exactly that against the recorded env.
  • src/pkg/ioscan/cli_client.go:191-198 — requiresModel is enforced in NewCLIChatClient, so an omp launch with an empty model is refused at construction rather than yielding "No default model selected" with exit 0 (which extractSingleJSONObject would otherwise report as a parse error on every kick). daily_quota_test.go:82-88 covers both the refusal and the accepted path.
  • src/cmd/hive/ioscan_classifier_boot.go:76-83,94-99 — the agents'-model fallback only engages for RequiresModel backends, so claude/copilot behaviour is unchanged; the ErrCLIModelRequired branch still sets UnavailableClassifier + alert at fail-closed before logging, same as the existing refusal path. ioscanClassifierAgentModel mirrors ioscanClassifierFallbackBackend (count, tie by name), so the choice is stable across boots.
  • Premise check: the hive process already reads OMP_AUTH_BROKER_URL from its own environment (src/crux/agent/manager_env.go:32), so os.LookupEnv in env() will find the broker pair where omp agents run.
  • reconcile_starvation_test.go — the waitReads gating is read-count based with a 10s deadline; newLoopClock is used by both tests and advance semantics are unchanged. Test-only, as the body says.

Open question (not a finding): I cannot verify omp's flag set (--no-tools --no-lsp --no-extensions --no-skills --no-rules --no-session --no-title --no-context-files --profile --system-prompt --mode text) from this repository — nothing else in the tree launches omp with these flags. The body says it was verified against a live binary; if any of them is version-gated, the launch fails loudly (nonzero exit → error → fail-closed alert) rather than silently running with tools, so the failure mode is safe.

go build ./pkg/ioscan/ ./cmd/hive/ passes at f2c10e7. The unrelated test deflake is declared in the body; scope is otherwise Closes #15873.

…mp agents

omp --profile isolates settings as well as auth, so the never-configured
hive-ioscan profile has no default model: launched without --model it prints
"No default model selected" with exit 0 and the classifier fails on every
call while looking configured. Boot now passes the model most of the hive's
omp agents run with when ioscan.classifier.model is empty, and refuses the
backend with one log line (and the blocked-kick alert at fail-closed) when
neither names one. claude and copilot keep their own CLI default.

Also records that omp reads OMP_AUTH_BROKER_URL/TOKEN from the environment
inside an isolated profile (verified against an unreachable broker), and
updates ioscan.md accordingly.

Hive-Fix-Attempt: 2/6
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
@hivecommons-hive hivecommons-hive Bot added stage/050-review-pending Review swarm requested; awaiting a verdict for the current head and removed stage/070-fixing Hive-authored PR: CI red, changes requested or needs rebase; a hive agent is fixing it labels Oct 31, 2026

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Read the diff against the v5 tree, the callers of NewCLIChatClient, and the settings validator. Looks correct to me.

What I verified:

  • src/pkg/ioscan/cli_client.go:85-105 — the omp spec reads the prompt on stdin (promptViaArg unset → Complete pipes it, cli_client.go:270-272), and argv() appends --model after the spec args (cli_client.go:228), so the launch matches the one documented in src/docs/ioscan.md.
  • cli_client.go:196-198 — requiresModel && model == "" returns a wrapped ErrCLIModelRequired; boot distinguishes it with errors.Is (src/cmd/hive/ioscan_classifier_boot.go:94) and still runs the fail-closed/fail-open branch first (:86-93), so the alert/UnavailableClassifier behaviour is unchanged for the new refusal.
  • ioscan_classifier_boot.go:176-201 — ioscanClassifierAgentModel iterates cfg.Agents with the same majority-then-name ordering as ioscanClassifierFallbackBackend (:148-169), so the backend and model fallbacks pick from the same population. Only consulted when cl.Model is empty and RequiresModel(backend); claude/copilot keep the pre-existing empty-model path.
  • src/pkg/dashboard/api_governor_security.go:128 validates via ioscan.SupportsToolless, so settings accept omp with no further change; the <select> in config-watchdog-security.js:836 is updated to match.
  • Env allowlist (cli_client.go:147) adds only the two broker vars; TestCLIChatClientOmpLaunchIsToolless asserts GH_TOKEN/GITHUB_TOKEN/HIVE_* are absent.

The reconcile_starvation_test.go deflake is unrelated to omp but is declared as such in the body; it is test-only.

— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive
hivecommons-hive Bot merged commit ca9557c into v5 Oct 31, 2026
79 of 82 checks passed
@kubestellar-prow
kubestellar-prow Bot deleted the scanner/fix-15873 branch October 31, 2026 01:20
@hivecommons-hive

Copy link
Copy Markdown
Contributor Author
Evidence bundle `#15888f2c10e7` sha256:25e7fc33eb54f892ce7fb6816f1a0f19b5a71895188757c5bf926db6452232ae — download from the dashboard or `hivectl review evidence #15888`.

@github-actions

Copy link
Copy Markdown
Contributor

Thank you for your contribution! Your PR has been merged.

We'd love to hear how your experience was: share feedback

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

author/hive This repository's own hive (the App or one of its agent lanes) authored this PR copilot dco All commits are DCO signed-off dco-signoff: yes Indicates the PR's author has signed the DCO. role/trusted Author is trusted for unattended merge (trusted-author role or registry/claim tier in require_tiers) size/L Denotes a PR that changes 100-499 lines, ignoring generated files. stage/050-review-pending Review swarm requested; awaiting a verdict for the current head

Projects

None yet

Development

Successfully merging this pull request may close these issues.

✨ ioscan: accept omp as a tool-less classifier CLI backend

0 participants