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.
npx skills add ChrisLamDev/hermes-core-skills --skill self-regulation-brake-systemgit clone --depth 1 https://github.com/ChrisLamDev/hermes-core-skillsWrote 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/skills/chrislamdev/hermes-core-skills/self-regulation-brake-system)<a href="https://agentmods.dev/skills/chrislamdev/hermes-core-skills/self-regulation-brake-system"><img src="https://agentmods.dev/badge/skills/chrislamdev/hermes-core-skills/self-regulation-brake-system/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/skills/chrislamdev/hermes-core-skills/self-regulation-brake-system"><img src="https://agentmods.dev/badge/skills/chrislamdev/hermes-core-skills/self-regulation-brake-system.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 | $0.00026 | $0.01649 |
| Opus 5 | $0.00013 | $0.00825 |
| Sonnet 5 | $0.00005 | $0.00330 |
| Haiku 4.5 | $0.00003 | $0.00165 |
Grade A, and why
self-regulation-brake-system 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 10d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **terminal 做到嘅**（開app、build、git、curl、cp/mv、改 code、compile）→ 用 terminal How it starts
The opening of the file, as written. The whole thing — 162 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Self-Regulation Brake System
Problem
Agent can burn thousands of tokens repeating the same failed action (GitHub login, OAuth, browser loops). The user only discovers this when they see the bill.
Solution: brake.sh
A lightweight shell script at ~/.hermes/scripts/brake.sh that:
- Tracks failure count per action name across sessions
- Exits with code 2 when limit (default: 3) is exceeded
- Resets on success
Integration Points
0. System-Level Injection（最強保證 — config.yaml prefill_messages_file）
Brake rules 已寫入 system instructions file，透過 config.yaml 嘅 prefill_messages_file 注入：
- System instructions file:
~/.hermes/system-instructions/self-regulation.md - 呢個 file 每次新 session 開波會自動 inject 做 system instruction
- 因為係 config-level injection，Agent 無法跳過或忽略
1. session-start.sh hook
Already updated. Every new session shows brake status:
=== 🧠 SELF-REGULATION CONTEXT ===
⚠️ BRAKE SYSTEM IS ACTIVE
2. Agent memory
Injected as self-regulation rules. Ensures the agent knows about brakes even after context compaction.
3. Manual usage (agent self-enforces)
Whenever the agent encounters a failing loop:
# Before retrying, check if brake is triggered
bash ~/.hermes/scripts/brake.sh check "github-login-by-browser"
# exit 0 = safe, count = N
# exit 2 = STOP, ask user
# After a failure
bash ~/.hermes/scripts/brake.sh record "github-login-by-browser"
# After success
bash ~/.hermes/scripts/brake.sh reset "github-login-by-browser"
Brake Trigger Rules（hard-coded in memory，亦注入 system instructions）
| Condition | Action |
|---|---|
| Same action fails ≥3 times | STOP → clarify user（用 clarify tool，唔好自己估） |
| 5 min of no progress | STOP → clarify user |
| Third-party auth (GitHub/Google/OAuth) | First failure → STOP, don't retry |
| Complex multi-step auth | Delegate to subagent first |
| 卡住咗唔出聲 | 違反 HARD RULE 2 — 失敗咗必須即刻報告，唔可以靜雞雞轉彎 |
🚩 CUA-Specific Auto-Stop Rules（2026-06-17 確立）
Terminal First 原則
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.
- 10d ago First seen · 162 lines · 26 tokens per session scan A 9d0259f2decc
self-regulation-brake-system is a skill published in the GitHub repository ChrisLamDev/hermes-core-skills (9 stars, last pushed 2mo ago), licensed MIT. It adds 26 tokens to every session and 1,649 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
unicli-repair
Evidence-driven repair workflow for a broken Uni-CLI adapter. Trigger on a failed unicli envelope, a quarantined adapter, or an explicit adapter-repair request. Classifies non-source failures, edits only the reported adapter path, and uses the original command as a bounded oracle.
dead-code-detector
Detect unused/unreachable code in polyglot codebases (Python, TypeScript, Rust). TRIGGERS - dead code, unused functions, unused imports.
multi-agent-performance-profiling
Multi-agent performance profiling for pipeline bottlenecks. TRIGGERS - performance profiling, bottleneck analysis, pipeline optimization.
bgclick-rev
Reverse-engineer a macOS GUI automation app's background-click path and produce a Swift reproduction. Use only when the user provides a target .app and explicitly asks for IDA-backed background-click reverse engineering. Requires IDA Pro with IDA MCP attached.
python-memory-safe-scripts
Memory-safe Python script patterns for long-running processes under systemd MemoryMax constraints. Covers allocator purge (mimalloc/glibc malloctrim), HTTP response lifecycle, DataFrame cleanup, thread-local connection reuse, and periodic GC cadence. Battle-tested through 5 OOM optimization cycles on production GPU…
tick-collection-ops
Operate and troubleshoot the MT5 tick collection system. HOST MOVED 2026-08-07 - MT5 now runs on the FXView VPS (Windows scheduled tasks, C:\odb-tickrings), NOT gpu-host-1/Linux-Wine/systemd. EA, ring format and gap-detection reasoning unchanged. Covers gap detection, restart recovery, daily ops.