known-issues

known-issues is a command for coding agents from ReviewToolkits/cpython-review-toolkit. It costs 43 tokens per session (1,657 once invoked), scanned A, original, MIT.

A command that compares a catalog of previously found CPython crashes with a new source-tree scan. A regression is a bug that returns after it was fixed.

In plain words
What is it for?
Use it to check selected files or the whole project against recorded recursion, error, memory, null-pointer, reference-counting, thread-race, and initialization issues.
Why use it?
It shows whether known problems are still present, have moved because the code changed, or appear to be fixed, without running reproduction programs.

Command

Part of the cpython-review-toolkit plugin — 7 commands, 23 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/reviewtoolkits/cpython-review-toolkit/known-issues
Clone the repo
git clone --depth 1 https://github.com/ReviewToolkits/cpython-review-toolkit

Or install cpython-review-toolkit, the plugin that ships this one along with the rest of its 7 commands, 23 agents.

Wrote 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.

agentmods badge for known-issues

README.md
[![agentmods](https://agentmods.dev/badge/commands/reviewtoolkits/cpython-review-toolkit/known-issues.svg)](https://agentmods.dev/commands/reviewtoolkits/cpython-review-toolkit/known-issues)
Your own site
<a href="https://agentmods.dev/commands/reviewtoolkits/cpython-review-toolkit/known-issues"><img src="https://agentmods.dev/badge/commands/reviewtoolkits/cpython-review-toolkit/known-issues.svg" alt="Measured on agentmods" height="20"></a>
Per session 43 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,657 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00043 $0.01657
Opus 5 $0.00022 $0.00829
Sonnet 5 $0.00009 $0.00331
Haiku 4.5 $0.00004 $0.00166

Measured 3d ago against content hash 351e6f71439d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

known-issues 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 3d ago.

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.

plugins/cpython-review-toolkit/commands/known-issues.md · 100 lines

How it starts

The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CPython Known-Issues Regression

data/cpython_known_bugs.tsv records crash sites previously confirmed in CPython — from the findings repos (OOM-####, TSAN-####) and the upstream tracker (gh-NNNNNN). This command runs, per catalog entry, the scanner matching that entry's category over the referenced file and reports whether the known bug is still in the tree.

It is static and drift-tolerant — it does not run repros. It answers: "of the crashes we already know about, which are still present at this checkout, which have drifted, and which look fixed?"

Scope: "$ARGUMENTS" (path or glob; default the whole project). The scope's checkout root is auto-detected — catalog paths are resolved relative to it, so a narrow scope still locates the right files.

Plugin root: <plugin_root> is the plugins/cpython-review-toolkit/ directory.

Category → scanner map

Category Scanner
recursion scan_recursion_guards.py
pyerr-clear scan_pyerr_clear.py
uninit-dealloc scan_uninit_dealloc.py
null-deref scan_null_checks.py
refcount scan_refcounts.py
tsan scan_ft_races.py
init-bypass scan_init_bypass.py

Every shipped category has a scanner as of v0.7 — no_scanner: 0 is the expected result, and a non-zero count means a category was added to the catalog before its scanner. Preserve that invariant when extending the catalog.

Workflow

  1. Cross-reference:

    python <plugin_root>/scripts/check_known_issues.py [scope]
    

    Optional flags: --max-files N, --catalog <path> (override the default TSV). Record the reviewed commit (git -C <project_root> rev-parse --short HEAD) — the catalog was captured at a slightly different commit, so line drift is expected.

  2. Read the catalog_results / bug_rollup blocks. Each entry gets one of:

    • present — a finding of that category at (or within ±5 lines of) the catalog line, or — when the catalog line is unknown (0) — a finding in the named function. Still unfixed. → high confidence.
    • line_drifted — the file still has findings of that category but not at the catalog site, and the catalog's named function is gone (renamed/removed) or was never recorded (0). nearest_line points at the closest finding. The bug likely just moved. → medium.
    • absent_in_function — the catalog names a function, that function still exists in the file, and it carries no finding of the category — while the file has findings elsewhere. This is absence in that function, not drift. Drift says "still there, moved"; this says "possibly fixed here". → weak signal, not a finding.
    • absent — the file was scanned and has no findings of that category anywhere (likely fixed or refactored).
    • file_missing — the path no longer exists under the checkout.
    • no_scanner — a catalog category with no entry in CATEGORY_SCANNERS; carried through for completeness, not cross-referenced. As of v0.7 every shipped category has a scanner, so this should read 0.

Read the full file on GitHub · 100 lines

Changes

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.

  1. 3d ago First seen · 100 lines · 43 tokens per session scan A 351e6f71439d

Subscribe to this mod's changes

known-issues is a command published in the GitHub repository ReviewToolkits/cpython-review-toolkit (10 stars, last pushed 1mo ago), licensed MIT. It adds 43 tokens to every session and 1,657 once invoked, about $0.0002 per session on Opus 5. 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 2026-08-31.