Skip to content
Agentic Cinematic Universe
Needs attention

2 things need attention.

Checked 2026-08-28, when this page was last built. See the whole org →

What needs you

spend·needs attention

No AI spend has been recorded, so the org cannot say what it costs to run

An empty log is not evidence of thrift. It means nothing measured the spend, and no routing decision can be checked against real usage.

What to do

Install the dispatch hook from hooks/README.md so SubagentStop and Stop write a row per dispatch. Then re-run npm run snapshot.

Owner machine-man·Source vault-schema/AI-HQ/Runs/routing.md
spend·needs attention

The telemetry hook has never fired on this machine

No session watermark exists, and dispatch-telemetry.sh writes one every time it runs. Nothing measured the spend here, and absent is not the same as zero.

What to do

Add the hooks from hooks/settings.snippet.json to your Claude Code settings, then check scripts/telemetry-coverage.sh.

Owner machine-man·Source hooks/.state

3 decisions only you can make

  • Confirm the canon is authoritative so the 14 legacy ledgers can be archived

    Waiting 0 days · asked by the-ancient-one · None. Legacy files stand as the check on the migration until confirmed

  • Set the spend threshold per venture for the risk matrix

    Waiting 0 days · asked by the-ancient-one · Every spend of any size is escalate until set. Fails closed, which is correct but noisy

  • Authorise Zoho CRM, Zoho Desk and Gmail connectors

    Waiting 0 days · asked by vision · new_paid_icp_accounts, mrr and revenue_concentration_top3 print UNKNOWN. The Strange decision due 2026-11-11 cannot be fully scored without them

See all 3, with what each one affects →

Searches workers, commands, procedures and the record at once. ⌘K opens it from any page.

Workers
127
Each one declares where its work lands and what finished means.
See who does what
Things you can ask for
142
No two give you the same thing back.
See the list
Cost so far
Unknown
Nothing measured it. That is not the same as it being cheap.
Self-checks

All 6 checks pass

The rules behind them →
  • Every agent says where its work lands, how often it runs, and what done means — passed127 of 127
  • No two commands produce the same thing — passed142 commands, 142 distinct outputs
  • Every command says what it produces — passed142 of 142
  • Every line in the shared record is readable — passed23 of 23
  • Every forecast says how it will be settled — passed1 of 1
  • Every agent's runs can be costed — passed127 of 127

Run against the files themselves every time this page is built, not asserted.