Borrowing it
Nothing to install: this file belongs to littlebearapps/untether. 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/littlebearapps/untether/master/.claude/commands/debug/systemic-patterns.mdgit clone --depth 1 https://github.com/littlebearapps/untetherWrote 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/littlebearapps/untether/systemic-patterns)<a href="https://agentmods.dev/commands/littlebearapps/untether/systemic-patterns"><img src="https://agentmods.dev/badge/commands/littlebearapps/untether/systemic-patterns/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/littlebearapps/untether/systemic-patterns"><img src="https://agentmods.dev/badge/commands/littlebearapps/untether/systemic-patterns.svg" alt="Reviewed on agentmods" width="80" 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 | $0.04377 |
| Opus 5.5 | $0.00000 | $0.01751 |
| Sonnet 5.5 | $0.00000 | $0.00875 |
| Haiku 4.5 | $0.00000 | $0.00438 |
Grade A, and why
systemic-patterns scanned grade A with 1 finding 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 18d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- **Sig**: subsequent Claude runs fail to spawn; `subprocess.create.failed` with FD-exhaustion-like errors; `lsof` shows growing FD count How it starts
The opening of the file, as written. The whole thing — 200 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Step 4 — Cross-reference systemic patterns
This is the canonical catalogue of known Untether systemic patterns. Walk this list before scoring any signal or escalating any "error" to a bug.
For each pattern:
- Sig = grep signature in journalctl / chat
- Class = primary Step-1 class
- Canonical issue = the open or closed GH issue (if any) tracking the pattern
- Posture =
bug(fix needed),by-design(do NOT escalate),regression-watch(was fixed, watch for relapse),rate-limited(real but expected at low frequency),mitigation-exists(Untether already mitigates this engine quirk)
The pattern catalogue
1. Auto-continue death spiral on signal exits
- Sig: rapid
auto_continuefollowed byproc_returncode=143or137, repeating - Class: auto-continue
- Canonical issue: mitigation via
_is_signal_death()inrunner_bridge.py; configurable[auto_continue].max_retries - Posture: regression-watch — signal-death suppression should kick in. If you see >1 retry on signal death, that's a bug.
2. PTY master_fd leak across runs
- Sig: subsequent Claude runs fail to spawn;
subprocess.create.failedwith FD-exhaustion-like errors;lsofshows growing FD count - Class: control-channel
- Canonical issue: per
control-channel.md§PTY lifecycle —finallymust closemaster_fd - Posture: regression-watch — every Claude runner exit path must close
master_fd.
3. callback_data > 64 bytes silently dropped
- Sig: button click does nothing; no
callback_queryin journalctl; outbound message had a button with longcallback_data - Class: telegram-transport
- Posture: bug if surfaced — Telegram enforces 64 bytes silently. Always check
callback_datalength at construction.
4. Hot-reload race during active run
- Sig:
config.reloadevent during an active session; subsequent events use stale config - Class: config-hot-reload
- Canonical:
telegram-transport.md§TelegramBridgeConfig hot-reload;update_from()must copy all fields atomically - Posture: regression-watch — new fields added to
TelegramBridgeConfigmust be added toupdate_from().
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.
- 18d ago Changed · +42 lines 23b34aaf5215
- 1mo ago First seen · 158 lines · 0 tokens per session scan A 1c8f966398f4
systemic-patterns is a command published in the GitHub repository littlebearapps/untether (74 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 4,377 tokens. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
root-cause
Use when any test fails, bug appears, or behaviour surprises you, before proposing a fix - find the cause and prove it, by reading real evidence, tracing bad values back to their origin, comparing against a working case, and testing one hypothesis at a time.
refactor
Refactor and optimize [code].
hotfix
Emergency fix workflow with audit trail.
kotlin-build
Fix Kotlin/Gradle build errors, compiler warnings, and dependency issues incrementally. Invokes the kotlin-build-resolver agent for minimal, surgical fixes.
bug-report
Create structured bug report.
wtfp:report-bug
Collect a reproducible bug report and publish it only after explicit user approval.