Shadow prompt loop
Run a new prompt in the shadows against real traffic alongside the current one; the disagreements decide whether to switch.
/loop cadence: continuous or per-batch. For real traffic [SOURCE], run the candidate prompt in shadow alongside the live prompt without exposing shadow output to users. Append disagreements (input, live output, shadow output, which was better by [CRITERIA]) to state-file shadow-prompt.md. Each round, summarize where they diverge and whether the candidate wins. Never serve shadow output to a user or change the live prompt automatically. Stop after the batch; log errors and stop. Budget: cap $[X]/run. Hard cap: stop after 1 iteration per run.
claude-code
Implementation note
ADAPTED, not verbatim. No tool named. Yellow loop: evaluation only; a human promotes the winner.
Source: EXM7777 ↗graded B · 75/100 — how grades work →
More research loops
Overnight intel refresh
A cheap model watches your tracked sources nightly; the expensive model writes only the monthly synthesis.
/loop cadence: nightly (synthesis monthly). Using Exa MCP + a Firecrawl monitor on tracked sources [LIST], have a cheap model log what materially changed to state-file intel-feed.md each night. Route to Fable only for the monthly synthesis: read the month's log and write ONE briefing of what changed and why it matters. Read + own-file writes only. Stop nightly after logging / monthly after the brief; log errors and stop. Budget: cheap-first, cap $[X]/run. Hard cap: stop after 1 iteration per run.
Kill-criteria loop
Make every live option declare the single fact that would disqualify it upfront, then hunt for that evidence.
/goal for each live option [LIST], state its ONE disqualifying condition upfront, then run an evidence hunt for that disqualifier. Done when: for every option, the state-file kill-criteria.md contains either PASTED evidence the disqualifier is true (kill it) or a documented search showing it isn't (keep it). No option stays undecided. Paste the evidence table as proof, or paste what's still missing and stop. Judge: a second smaller model checks each verdict cites evidence. Budget: cap $[X]. Hard cap: stop after 1 iteration per run.
Pre-mortem loop
Before committing, write the 'it's 12 months later and this failed' story so the failure modes are on the table now.
/goal for decision [X], write the pre-mortem: assume it's 12 months later and this failed, then enumerate the specific causes, ranked by likelihood, each with an early-warning signal and a mitigation. Done when: state-file premortem.md contains the failure narrative + a ranked cause table with signals and mitigations, and the top 3 causes each have a concrete owner/mitigation PASTED in. Paste the completed table as proof, or paste what's unfinished and stop. Budget: cap $[X]. Hard cap: stop after 1 iteration per run.