Skip to content

akiiwa-games/AgentCompass / skills/agent-compass

agent-compass

Flagged

Choose owner, model and effort, outbound authorization, advisor use, and continuation for delegated work. Reads local usage and effective account policy, profiles the task with the configured System-One scorer, then allocates a compatible worker. Use for independent worker or advisor work; ordinary local execution needs no routing.

FileKindSize
SKILL.mdskill md5.5 KB
agents/openai.yamlconfig427 B
calibration/README.mdmarkdown55.2 KB
calibration/arena_capture.pyscript14.0 KB
calibration/community.pyscript25.2 KB
calibration/five_axis.pyscript196.4 KB
calibration/measurements.pyscript68.4 KB
calibration/refresh.pyscript48.6 KB
calibration/series.pyscript10.2 KB
calibration/sources.pyscript24.1 KB
calibration/tiers.pyscript14.6 KB
grok-bot/.gitignoreother47 B
grok-bot/dependency_patch.pyscript4.5 KB
grok-bot/install_cli.pyscript3.6 KB
grok-bot/test-recovery.mjsscript8.6 KB
grok-bot/test_install_cli.pyscript4.1 KB
grok-bot/upstream.jsonconfig3.4 KB
references/accounts.jsonconfig155.9 KB
references/classifier-cases.jsonconfig21.5 KB
references/dispatch-lifecycle.mdmarkdown28.9 KB
references/dispatch-rules.mdmarkdown11.3 KB
references/effort-matching.mdmarkdown5.1 KB
references/fast-modes.mdmarkdown3.5 KB
references/five-axis-scores.jsonconfig3.1 MB
references/gemini-decisions-api.mdmarkdown2.9 KB
references/grok-bot.mdmarkdown13.0 KB
references/herdr.mdmarkdown5.1 KB
references/jev-api.mdmarkdown1.6 KB
references/level-3-conversion.jsonconfig2.4 KB
references/model-capabilities.jsonconfig3.6 KB
references/model-capabilities.mdmarkdown14.6 KB
references/parent-address-gate.mdmarkdown10.1 KB
references/plan-templates.jsonconfig13.2 KB
references/provider-discovery.mdmarkdown11.5 KB
references/provider-usage.mdmarkdown4.1 KB
references/public-data.jsonconfig816 B
references/public-measurements.jsonconfig1.3 MB
references/roll-reference.mdmarkdown26.7 KB
references/router-internals.mdmarkdown93.7 KB
references/task-classifier.mdmarkdown6.1 KB
references/user-scores.jsonconfig6.8 KB
references/work-kinds.mdmarkdown4.7 KB
references/worker-communication.mdmarkdown2.1 KB
scripts/account_fleet.pyscript32.9 KB
scripts/account_state.pyscript22.2 KB
scripts/accounts_override.pyscript14.8 KB
scripts/allocator.pyscript19.8 KB
scripts/antigravity_plan.pyscript18.7 KB
scripts/antigravity_telemetry.pyscript5.1 KB
scripts/auto_bundles.pyscript31.3 KB
scripts/axis_scores.pyscript40.3 KB
scripts/caller_input.pyscript3.0 KB
scripts/candidate_eval.pyscript54.0 KB
scripts/candidate_ranking.pyscript9.5 KB
scripts/capability_catalog.pyscript8.0 KB
scripts/check_router_contract.pyscript2.8 KB
scripts/conftest.pyscript2.5 KB
scripts/cwd_policy.pyscript1.5 KB
scripts/data_age.pyscript1.5 KB
scripts/decision_ledger.pyscript18.9 KB
scripts/demo_plan.pyscript6.3 KB
scripts/dispatch_fallback.pyscript40.3 KB
scripts/dispatch_flow.pyscript58.7 KB
scripts/dispatch_ground_truth.pyscript43.5 KB
scripts/dispatch_registry.pyscript113.7 KB
scripts/entry.pyscript1.7 KB
scripts/env_vars.pyscript580 B
scripts/fast_modes.pyscript2.5 KB
scripts/fixtures/grok-api-responses.jsonconfig907 B
scripts/fixtures/reference-accounts.jsonconfig26.9 KB
scripts/grok_bot_paths.pyscript1005 B
scripts/host_herdr.pyscript16.8 KB
scripts/host_placement.pyscript9.4 KB
scripts/host_policy.pyscript10.1 KB
scripts/launch_failures.pyscript5.9 KB
scripts/launch_values.pyscript12.3 KB
scripts/ledger_snapshot.pyscript11.1 KB
scripts/live_models.pyscript21.1 KB
scripts/model_latch.pyscript7.6 KB
scripts/model_resolution.pyscript18.0 KB
scripts/override_store.pyscript2.2 KB
scripts/plan_state_residue.pyscript5.0 KB
scripts/provider_inventory.pyscript128.2 KB
scripts/provider_retry.pyscript34.6 KB
scripts/public_data.pyscript59.2 KB
scripts/quota_economy.pyscript38.0 KB
scripts/quota_exhaustion.pyscript5.9 KB
scripts/quota_overrides.pyscript25.6 KB
scripts/quota_pools.pyscript29.8 KB
scripts/reference_provenance.pyscript32.3 KB
scripts/released_names.pyscript3.8 KB
scripts/roll.pyscript106.3 KB
scripts/route_judge.pyscript83.4 KB
scripts/route_simulation.pyscript26.3 KB
scripts/router_test_support.pyscript6.7 KB
scripts/run_grok_bot.pyscript18.8 KB
scripts/scorer_client.pyscript55.2 KB
scripts/stage_help.pyscript5.8 KB
scripts/state_upgrade.pyscript24.0 KB
scripts/subscription_cli.pyscript3.6 KB
scripts/task_intake.pyscript33.6 KB
scripts/test_account_identity.pyscript7.1 KB
scripts/test_account_state.pyscript63.3 KB
scripts/test_accounts_override.pyscript30.1 KB
scripts/test_allocator.pyscript94.2 KB
scripts/test_antigravity_plan.pyscript13.2 KB
scripts/test_applicable_quota.pyscript9.1 KB
scripts/test_auto_bundles.pyscript10.8 KB
scripts/test_billing_routing_policy.pyscript9.9 KB
scripts/test_cadence_count.pyscript32.2 KB
scripts/test_calibration_units.pyscript27.0 KB
scripts/test_capability_catalog.pyscript5.0 KB
scripts/test_decision_ledger.pyscript8.5 KB
scripts/test_demo_plan.pyscript3.4 KB
scripts/test_derived_state.pyscript21.6 KB
scripts/test_dispatch.pyscript322.3 KB
scripts/test_dispatch_answers.pyscript14.8 KB
scripts/test_dispatch_fallback.pyscript53.2 KB
scripts/test_entry.pyscript1.6 KB
scripts/test_fast_modes.pyscript9.9 KB
scripts/test_five_axis.pyscript290.1 KB
scripts/test_five_axis_economy.pyscript105.0 KB
scripts/test_fleet_parity.pyscript15.6 KB
scripts/test_grok_bot.pyscript47.7 KB
scripts/test_ground_truth_recording.pyscript68.8 KB
scripts/test_herdr.pyscript14.1 KB
scripts/test_host_policy.pyscript24.6 KB
scripts/test_launch_values.pyscript8.0 KB
scripts/test_metered_usd.pyscript60.4 KB
scripts/test_model_resolution.pyscript9.2 KB
scripts/test_named_route_facts.pyscript28.6 KB
scripts/test_native_hosts.pyscript6.0 KB
scripts/test_new_user_fleet.pyscript3.3 KB
scripts/test_no_hardcode_classification.pyscript75.1 KB
scripts/test_no_legacy_scoring.pyscript6.5 KB
scripts/test_override_store.pyscript2.3 KB
scripts/test_plan_says_why.pyscript44.9 KB
scripts/test_provider_inventory.pyscript83.2 KB
scripts/test_provider_names.pyscript126.0 KB
scripts/test_provider_pools.pyscript14.2 KB
scripts/test_public_data.pyscript93.5 KB
scripts/test_quota_exhaustion.pyscript5.6 KB
scripts/test_rate_limit_windows.pyscript6.3 KB
scripts/test_record_says_why.pyscript26.5 KB
scripts/test_reference_provenance.pyscript16.7 KB
scripts/test_refresh_entry.pyscript110.7 KB
scripts/test_released_names.pyscript4.1 KB
scripts/test_reset_priority.pyscript4.8 KB
scripts/test_roll_answers.pyscript44.7 KB
scripts/test_roll_notifications.pyscript10.4 KB
scripts/test_route_judge.pyscript99.8 KB
scripts/test_route_simulation.pyscript7.0 KB
scripts/test_routing_parity.pyscript32.3 KB
scripts/test_routing_review.pyscript25.1 KB
scripts/test_setup_discovery_parity.pyscript9.5 KB
scripts/test_stage_help.pyscript2.0 KB
scripts/test_state_upgrade.pyscript13.8 KB
scripts/test_turn_refusal.pyscript16.7 KB
scripts/test_usage_cache.pyscript4.9 KB
scripts/test_usage_collector.pyscript21.4 KB
scripts/test_usage_core_claude_accounts.pyscript1.8 KB
scripts/test_usage_core_grok_cli_billing.pyscript15.6 KB
scripts/test_usage_core_magpie.pyscript5.4 KB
scripts/test_usage_core_magpie_pool.pyscript10.8 KB
scripts/test_usage_core_safety.pyscript24.4 KB
scripts/test_user_scores.pyscript102.3 KB
scripts/test_venice_usage.pyscript1.3 KB
scripts/test_window_scope.pyscript5.5 KB
scripts/turn_refusal.pyscript10.7 KB
scripts/usage_cache.pyscript4.7 KB
scripts/usage_collector.pyscript7.4 KB
scripts/usage_core.pyscript142.0 KB
scripts/usage_magpie.pyscript10.6 KB
scripts/usage_reading.pyscript17.1 KB
scripts/user_scores.pyscript72.8 KB
scripts/venice_usage.pyscript2.6 KB
scripts/window_scope.pyscript3.2 KB
scripts/worker_context.pyscript8.9 KB
scripts/worker_sessions.pyscript10.9 KB

scripts/candidate_ranking.py

View on GitHub
"""Ranking and ordering of evaluated candidates."""


import host_policy
import launch_failures
import live_models
import model_latch
import provider_retry
import account_fleet
import axis_scores
import candidate_eval
import quota_economy


def five_axis_order(candidate):
    """Lexicographic order on the five-axis path (quota economy v3, owner decision of round 13b + R2-F4), in tuple order:
    1. Net score N = F - cost descending (exact float, never rounded: the quota economy C is a term of N, never a tie-break)
    2. Wait before the start ascending (a candidate that waits for a reset trails one that starts now at equal N)
    3. Faster completion time (wait_hours + expected_hours) ascending; speed breaks exact ties on N (including C ≈ 0 near reset)
    4. Fewer list-price USD spent ascending
    5. Demand-weighted category level descending (five_axis_level), then sibling harness place ascending."""
    k = candidate['components']
    wait_h = float(candidate.get('available_in_hours') or 0.0)
    done_key = round(wait_h + float(candidate['duration_hours']), 4)
    usd = float(candidate.get('expected_usd') if candidate.get('expected_usd') is not None else float('inf'))
    return (-float(candidate['score']), round(wait_h, 4), done_key,
            round(usd, 4), -round(float(k.get('five_axis_level', 0.0)), 4), k.get('harness_place', 0))


FIVE_AXIS_ORDER_KEYS = ('net_score', 'wait_hours', 'completion_hours', 'usd', 'axis_level', 'harness_place')


def five_axis_deciding_key(first, second):
    """The name of the first key of `five_axis_order` on which `first` beats `second` ('tie' when none does)."""
    for name, a, b in zip(FIVE_AXIS_ORDER_KEYS, five_axis_order(first), five_axis_order(second)):
        if a != b:
            return name
    return 'tie'


def candidate_for_bundle(rows, bundle_id, fast=False):
    """The ranking row for a bundle id and a fast flag: `<id>` is the plain variant, `<id>+fast` the fast one, and a fast
    flag with a plain id is the fast variant of that bundle. None when there is no such row."""
    for c in rows:
        if c['id'] == bundle_id and (not fast or c['bundle'].get('fast')):
            return c
    return next((c for c in rows if c.get('base_id') == bundle_id and bool(c['bundle'].get('fast')) == bool(fast)), None)


def rank(task, usage, policy, accounts, capabilities, now, unavailable=(), overrides=(), records=(),
         host_policy_config=None, *, read_local_state=True, named_accounts=()):
    state_policy = (host_policy.load_policy() if read_local_state else {'allowed_accounts': list((accounts.get('accounts') or {}).keys())}) if host_policy_config is None else host_policy_config
    ctx = {'policy': policy, 'accounts': accounts, 'capabilities': capabilities, 'usage': usage, 'now': now,
           'unavailable': set(unavailable), 'bundles': account_fleet.bundles(accounts),
           'host_policy': state_policy,
           'overrides': list(overrides), 'records': list(records or ()),
           'launch_failures': launch_failures.active() if (read_local_state and launch_failures.enabled()) else {},
           'model_latches': model_latch.active(now=now) if read_local_state else {},
           'provider_retries': provider_retry.active(now=now) if read_local_state else {},
           # A fact is read for every account a route may take: those in automatic routing and the ones a named
           # route or an explicit resume names, whose live list is otherwise never fetched.
           'live_models': live_models.load_or_refresh(
               accounts=accounts,
               allowed=lambda name: host_policy.account_allowed(state_policy, name) or name in named_accounts,
               named=named_accounts,
               now=now,
           ) if read_local_state else {'accounts': {}}}
    pool_bundles = dict(ctx['bundles'])         # the catalog, before a task-named effort adds bundles to it
    ctx['bundles'].update(candidate_eval.named_effort_bundles(task, ctx))
    ctx['_pin_matched_ids'] = candidate_eval.pin_matched_ids(task, ctx)
    five = ctx['five_axis'] = axis_scores.five_axis_config(policy)
    demand = axis_scores.five_axis_demand(task)
    panel_targets = accounts['_five_axis']['targets']
    task = {**task, 'five_axis_demand': demand, 'five_axis_targets': axis_scores.five_axis_targets(demand, panel_targets),
            'five_axis_continuity': axis_scores.five_axis_continuity_value(demand, panel_targets)}
    candidates = [candidate_eval.evaluate_bundle(b, task, ctx) for b in ctx['bundles'].values()]
    # R2-F7: when `--model` is named without `--effort` (and no bundle is pinned), never let an unscored effort beat a
    # scored effort of the same model via a silent 0.0 default. Only when ALL eligible candidates lack trusted scores
    # do they proceed together as an explicit `unscored_named_dispatch` with `fit = None`.
    if task.get('constraint_model') and not task.get('constraint_effort') and not task.get('pinned_bundle') and not task.get('constraint_bundle'):
        elig_now = [c for c in candidates if not c['excluded']]
        if any(not c.get('_five_blind_msg') for c in elig_now):
            shown_model = task.get('constraint_raw_model') or task['constraint_model']
            for c in elig_now:
                msg = c.get('_five_blind_msg')
                if msg:
                    blind_note = f'data error: {msg}; dispatched because it was named'
                    c['notes'] = [n for n in c['notes'] if n != blind_note]
                    if 'data_errors' in c:
                        c['data_errors'] = [e for e in c['data_errors'] if e != msg]
                        if not c['data_errors']:
                            c.pop('data_errors', None)
                    c['excluded'].append(
                        f'data error: {msg} (excluded because another effort of {shown_model} has trusted data)'
                    )
    for c in candidates:
        if not c['excluded'] and c.get('_five_blind_msg'):
            c['components']['fit'] = None
            c['components']['unscored_named_dispatch'] = True
            c['unscored_named_dispatch'] = True
    eligible = [c for c in candidates if not c['excluded']]
    pool_task = {**task, 'local_only': False, 'outbound_ok': None, 'file_transfer': False,
                 'needs_visual': False, 'needs_gui': False, 'wants_fast': False,
                 'read_only_review': False, 'constraint_bundle': None, 'constraint_raw_model': None,
                 'constraint_model': None, 'constraint_effort': None, 'pinned_bundle': None, 'user_named_bundle': None,
                 'metered_ok': True, 'metered_fallback': False, 'disable_fast': False, 'force_fast': False,
                 '_economy_pool': True}
    pool = [c for c in (candidate_eval.evaluate_bundle(b, pool_task, ctx) for b in pool_bundles.values()) if not c['excluded']]
    quota_windows = quota_economy.apply_five_axis_economy(candidates, eligible, task, accounts, policy, pool=pool)
    # Net score N = fit + continuity (both in panel standard deviations) + a temporary boost - the economy's cost (already
    # in score units). No normalization over the candidate set.
    weights = dict(five['weights'])
    kind = task['kind']
    for c in eligible:
        raw_score = (sum(weights[k] * (0.0 if c['components'].get(k) is None else c['components'][k]) for k in weights)
                     + c['components'].get('boost', 0)
                     - c['components']['opportunity_cost'])
        # Evaluate at IEEE-754 unit scale so F = 0 has the same ~2.2e-16 float64 resolution as F > 0 (shift invariance).
        c['score'] = float((1.0 + raw_score) - 1.0)
    # Equal scores go to the candidate with no wait, then faster completion, then fewer USD,
    # then higher level; the harness order only separates siblings of one model, account and effort.
    harness_places(eligible, kind, policy, capabilities)
    eligible.sort(key=five_axis_order)
    for i, c in enumerate(eligible):
        c['components']['decided_by'] = (five_axis_deciding_key(c, eligible[i + 1]) if i + 1 < len(eligible)
                                         else 'only_candidate')
    system = {'weights': {k: round(v, 3) for k, v in weights.items()}, '_live_models': ctx['live_models'],
              'quota_windows': quota_windows}
    return candidates, eligible, system


def sibling_key(candidate):
    """Bundles that differ only in harness (Codex or Pi running the same model on the same account)."""
    bundle = candidate['bundle']
    return bundle['account'], bundle['model'], bundle.get('effort')


def harness_group(candidate):
    """Siblings that would also do the same thing: resuming the owner is not the same choice as starting its twin."""
    return sibling_key(candidate) + (candidate['action'],)


def harness_places(eligible, kind, policy, capabilities):
    """Five-axis path: each candidate's place behind its best sibling (same account, model, effort and action) in the
    harness order `harness_places` computes, as `harness_place`; the score is not touched."""
    order = (capabilities['routes'].get(kind) or {}).get('harness_order') or policy.get('harness_order') or []

    def place(candidate):
        b = candidate['bundle']
        harness = b.get('harness') or account_fleet.bundle_harness(b, b.get('account'), b.get('id'))
        return order.index(harness) if harness in order else order.index(b.get('provider')) if b.get('provider') in order else len(order)

    groups = {}
    for c in eligible:
        groups.setdefault(harness_group(c), []).append(c)
    for siblings in groups.values():
        best = min(place(c) for c in siblings)
        for c in siblings:
            c['components']['harness_place'] = place(c) - best
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall