Mariomarquezt/multi-flow-skills / skills/sliced-review
sliced-review
FlaggedReview a whole codebase with several models by cutting it into small file-listed slices ordered by blast radius, then triage every finding with three judges, and optionally fix. Use when the user asks to review or audit a whole project or repo with other models, wants a second opinion on a codebase, wants a review split into slices or tiers, has a whole-repo review that died on credits, or wants a check before a launch or handover. Modes (`3`, `5`, `8`, `fast`, `fix …`, `fix fast …`, `fix only`, `report`, `no wrap`) are in the body. Not for a plan before code exists (/grill-me-codex, /codex-review) or a single diff, branch or PR (/code-review).
templates/scorecard.md
View on GitHub# Model scorecard — <project> — <YYYY-MM-DD>
Mode: `<invocation>` · slices: <N> · reviewers: <N> · triage done by: <who>
Scores use findings that SURVIVED triage (Step 10), never raw claims.
Weights: critical 8 · high 4 · medium 2 · low 1 · info 0.
Score = Σ weights of verified findings. Precision = verified ÷ reported.
Unique = verified findings no other model reported. Missed = verified critical/high found by others, not by this model.
| rank | model | route · effort | reported | verified | refuted | precision | crit | high | med | low | unique | missed crit/high | score | total time | avg / slice | slowest slice | cost | tokens | api_usd |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | <model> | <route> · <effort> | 0 | 0 | 0 | 0% | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0h 00m | 0m 00s | 0m 00s (<slice id>) | <ledger> | in <n> · cached <n> · out <n> | $<n> or unknown |
| 2 | <model> | <route> · <effort> | 0 | 0 | 0 | 0% | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0h 00m | 0m 00s | 0m 00s (<slice id>) | <ledger> | in <n> · cached <n> · out <n> | $<n> or unknown |
Severity columns count VERIFIED findings only. Refuted = disproved in Step 10 or 11b.
Time columns come from `ledger.tsv`: total = sum of that model's slice wall clocks; avg = total ÷ slices it ran; slowest = its longest single slice. Time is queue time per model, not the run's overall wall clock — lanes run in parallel.
`cost` = what the backend reported or billed (blank on a flat plan). `tokens` + `api_usd` come from the ledger's last two columns (usage.py): api_usd is the API-equivalent price of those tokens, filled for EVERY model — flat plan and free routes included — so models are compared on one scale. `unknown` = no price found (never estimated).
## Why the ranking
- **#1 <model>** — one sentence: what it found that mattered, and what it missed.
- **#2 <model>** — same.
## Every model missed
- <the verified critical/high that NO reviewer reported, found by the human or the database in triage — or "none">
## Caveats
- Ties on score break by unique finds, then precision.
- <N> findings were refuted after being reported by more than one model — agreement was not evidence.
- <anything that makes this run not comparable to another: partial tiers, a lane that died, buffered output kill, different slice set>
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |