Recursive Decision Ledger
- 1.4k installs
- 238k repo stars
- Updated August 5, 2026
- affaan-m/everything-claude-code
recursive-decision-ledger is a Claude Code skill that turns repeated agent rollouts and recursive prompting into an auditable append-only decision ledger with promotion gates.
About
recursive-decision-ledger is a Claude Code skill for developers forcing deeper computation through repeated rollouts or Prime Gauss style recursive prompting. It preserves repeated trials, prior memory, fresh information, and explicit marks while removing false certainty, recording each rollout under a ledger contract (rollout id, prior winner, fresh info, search space size, trial counts, decision and coherence marks). It prefers append-only JSONL ledgers with Markdown summaries. Developers use it when building agents that compare ensemble outputs, explore local optima, or run stochastic optimization and need an auditable decision log.
- Auditable ledger for repeated rollouts / recursive prompting instead of opaque loops
- Ledger contract records rollout id, prior winner, fresh info, search space, trial counts, decision marks
- Promotion rules default to paper/dry-run and block live actions without explicit approval and gates
Recursive Decision Ledger by the numbers
- 1,399 all-time installs (skills.sh)
- +216 installs in the week ending Aug 5, 2026 (Skillselion tracking)
- Ranked #845 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
recursive-decision-ledger capabilities & compatibility
- Capabilities
- orchestration · research · decision logging
- Use cases
- orchestration · research · memory
What recursive-decision-ledger says it does
tools: Read, Write, Edit, Bash, Grep, Glob
Prefer JSONL for append-only ledgers and Markdown for human summaries.
recursive confidence is not approval.
npx skills add https://github.com/affaan-m/everything-claude-code --skill recursive-decision-ledgerAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 1.4k |
|---|---|
| repo stars | ★ 238k |
| Last updated | August 5, 2026 |
| Repository | affaan-m/everything-claude-code ↗ |
What it does
Keep an auditable append-only ledger of repeated agent rollouts and ensemble comparisons, with promotion gates that block live actions without explicit approval.
Who is it for?
Agents that compare ensemble outputs, explore local optima, or run stochastic optimization and need an auditable log.
Skip if: Treating recursive confidence as approval for trading, capital allocation, production deploys, or destructive ops.
When should I use this skill?
The user asks for repeated rollouts, marked decision processes, stochastic optimization, or recursive reasoning with a visible evidence trail.
What you get
An append-only ledger of rollouts with decision and coherence marks and explicit promotion gates.
- append-only JSONL ledger
- Markdown summary
- coherence mark
By the numbers
- seven-step rollout loop
- ten ledger-contract fields
Files
Recursive Decision Ledger
Use this skill when the user is trying to force deeper computation through repeated rollouts or "Prime Gauss" style recursive prompting. Preserve the useful part: repeated trials, prior memory, fresh information, and explicit marks. Remove the unsafe part: pretending the loop proves certainty.
Ledger Contract
Every rollout should record:
- rollout id and timestamp;
- prior accepted winner and prior watchlist;
- fresh information ingested;
- search space size;
- model families or heuristics used;
- trial count and effective trial count;
- top candidates;
- decision marks;
- coherence marks against the prior ledger;
- promotion gate result.
Prefer JSONL for append-only ledgers and Markdown for human summaries.
Rollout Loop
1. Load the prior ledger. 2. Capture new information at time-step zero. 3. Run the bounded search. 4. Mark each candidate: accept, watch, reject, decay watch, or needs replay. 5. Compare winners against prior winners and latest marked rollout. 6. Downgrade candidates when drift, tail risk, stale data, or failed replay invalidates the previous mark. 7. Append artifacts before summarizing.
Coherence Mark
Include a compact coherence mark:
Ensemble matches prior winner: true
Recursive matches prior winner: false
Latest rollout match: true
Live promotion allowed: false
Reason: replay and freshness gates not satisfiedPromotion Rules
For trading, capital allocation, production deploys, migrations, or destructive ops, recursive confidence is not approval.
Default to paper, dry-run, read-only, preview, or staged mode unless the user explicitly approves the live action and the repo/service gate supports it.
Promote only when:
- the candidate beats the prior accepted winner on the chosen metric;
- correctness and replay checks pass;
- risk limits are explicit;
- the evidence is durable;
- the user has approved the live step when needed.
Summary Shape
Lead with the decision, not the drama:
Rollout 15 complete. The prior winner still holds, but edge deteriorated 17%.
Status: watch, not live. Next gate: 20 replay fills with fresh orderbook age
below threshold.Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10FAQ
What format is the ledger?
JSONL for append-only ledgers and Markdown for human summaries.
Can it approve a live action?
No. Recursive confidence is not approval; it defaults to paper/dry-run/read-only unless the user explicitly approves and the gate supports it.