Borrowing it
Nothing to install: this file belongs to stevologic/security-recipes.ai. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/stevologic/security-recipes.ai/main/.github/prompts/leftover-review.mdgit clone --depth 1 https://github.com/stevologic/security-recipes.aiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/stevologic/security-recipes.ai/leftover-review)<a href="https://agentmods.dev/commands/stevologic/security-recipes.ai/leftover-review"><img src="https://agentmods.dev/badge/commands/stevologic/security-recipes.ai/leftover-review.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.01051 |
| Opus 5 | $0.00000 | $0.00526 |
| Sonnet 5 | $0.00000 | $0.00210 |
| Haiku 4.5 | $0.00000 | $0.00105 |
Grade A, and why
leftover-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the daily leftover-gold reviewer for security-recipes.ai.
Live-verify and rewrite the leftover-gold CVE pages selected by
python scripts/pick_leftover_review_queue.py --json. The picker
drains leftover-gold criticals and highs first, then takes up to
25 leftover-gold medium and low pages. Do not invent a different
queue.
In scope:
- leftover-titled pages listed by the picker;
- data/cve/leftover-review-state.json reviewed_cves updates;
- markdown-only CVE catalog sync and dependent generated artifacts required by those page edits.
Out of scope:
- promoting leftover pages to maturity: stable without a verified single named floor;
- skip-family leftover dumps (Linux kernel, Firefox, Safari, iOS, Thunderbird, SourceCodester, Tenda, Wavlink, Edimax, D-Link, QNAP, Samsung, ZKTeco, Cisco Catalyst, Open Babel, FreeRDP, Locutus, and the other picker SKIP_RE families) unless the picker selected them;
- speculative changes, cosmetic churn, and date-only edits;
- weakening validation or modifying unrelated application code.
Work the review end to end:
- Start from the current default branch. Inspect recent leftover
review PRs with
gh pr list --state all --search "label:automation:leftover-review"so work is not duplicated. - Run
python scripts/pick_leftover_review_queue.py --jsonand review only that selected list. If selected_total is 0, append a concise explanation to$GITHUB_STEP_SUMMARYand finish without creating a branch, issue, or PR. - For each selected CVE, fetch live GHAD
https://api.github.com/advisories/<GHSA>withAuthorization: Bearerfrom GITHUB_TOKEN/GH_TOKEN, and live NVDhttps://services.nvd.nist.gov/rest/json/cves/2.0?cveId=. Never print tokens. If GHAD is 404, fetch the repository advisory. If NVD has no row, retry once, then say NVD has no row. Do not invent CPE. - Rewrite each leftover page in leftover gold style: stay development; set lastmod to today's UTC date (the review day), not a hardcoded leftover date; descriptions 110-155 characters; honesty tokens required (first_patched / no public floor / vulnerabilities empty / do not invent / live GHAD 404); do not write "GHAD names", "per GHSA", or "GHAD-named" as floor claims; never invent +1 versions; keep product-family leftover IDs distinct; add rollback / do not roll back into the affected range; do not copy GHSA or issue PoCs/exploits.
- Append every completed CVE ID to data/cve/leftover-review-state.json reviewed_cves, keep the list sorted and unique, and do not drop existing IDs.
- Run
python scripts/sync_cve_catalog.py --markdown-onlyandpython scripts/validate_cve_catalog.py. Hub counts must stay 33 stable / 24 AI-qualified / 57 indexable unless a catalog sync intentionally changes them. Then runnode --test tests/test_recipe_quality.js tests/test_search_intent.jsandpython -m unittest tests.test_cve_landing tests.test_cve_recipe_catalog tests.test_compliance_catalog tests.test_code_hygiene_catalog tests.test_leftover_review_queue tests.test_leftover_review_workflow. If content hashes change, runpython scripts/run_generator_pipeline.py --write, restore CRLF-only noise in scripts/generated-output-ownership.json and data/evidence/workflow-control-plane-report.json, then runpython -m unittest tests.test_generator_pipeline.GeneratorPipelineTests.test_all_generated_artifacts_are_fresh. - If no selected leftover page can be honestly rewritten, append
a concise explanation to
$GITHUB_STEP_SUMMARYand finish without creating a branch, issue, or PR. - If changes are justified, create a unique branch named
automation/leftover-review-, commit, push, and
open a PR to main. Explain the live GHAD/NVD evidence, floors,
and verification. Apply the label automation:leftover-review
(create it if necessary), then enable auto-merge with
gh pr merge --auto --squash <pr-number>. Never push directly to main or merge directly. Do not commit artifacts/, tokens, or exploit PoCs.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today First seen · 80 lines · 0 tokens per session scan A bff0a26470fc
leftover-review is a command published in the GitHub repository stevologic/security-recipes.ai (1 stars, last pushed yesterday), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,051 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-07.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.