https://github.com/88plug/trigger-my-training
Ground-first reflex for Claude Code: the agent judges complexity from its own training, grounds before acting, and is hard-blocked from the irreversible step until it has.
https://github.com/88plug/trigger-my-training
agent-safety ai ai-agents anthropic automation claude claude-code claude-code-plugin claude-skills cli developer-tools grounding guardrails infrastructure irreversible-actions llm preflight productivity reasoning safety
Last synced: 5 days ago
JSON representation
Ground-first reflex for Claude Code: the agent judges complexity from its own training, grounds before acting, and is hard-blocked from the irreversible step until it has.
- Host: GitHub
- URL: https://github.com/88plug/trigger-my-training
- Owner: 88plug
- License: other
- Created: 2026-06-21T23:06:58.000Z (about 1 month ago)
- Default Branch: main
- Last Pushed: 2026-07-17T14:44:49.000Z (7 days ago)
- Last Synced: 2026-07-17T15:22:28.916Z (7 days ago)
- Topics: agent-safety, ai, ai-agents, anthropic, automation, claude, claude-code, claude-code-plugin, claude-skills, cli, developer-tools, grounding, guardrails, infrastructure, irreversible-actions, llm, preflight, productivity, reasoning, safety
- Language: Python
- Size: 147 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING.md
- License: LICENSE.md
- Security: SECURITY.md
Awesome Lists containing this project
- awesome-claude-skills - 88plug/trigger-my-training
README
# Trigger My Training
**Ground-first reflex for Claude Code & Grok** — agent safety and grounding.
The agent judges complexity from its own training, grounds before acting, and
is hard-blocked from irreversible steps until it has.
[](https://github.com/88plug/trigger-my-training/actions/workflows/plugin-validate.yml)
[](LICENSE)
[](https://88plug.github.io/trigger-my-training/)
[](https://github.com/88plug/claude-code-plugins)
[](https://deepwiki.com/88plug/trigger-my-training)
> [!IMPORTANT]
> The core insight, after an adversarial review that killed the naive version:
> **an agent's training is always stale.** The reflex does not use recall as the
> *answer* — it uses recall to generate the *list of things to verify*, then
> runtime overrides recall. That reframing is what makes grounding help instead
> of hurt.
## Install
### Claude Code
```text
/plugin marketplace add 88plug/claude-code-plugins
/plugin install trigger-my-training@88plug
```
### Grok Build
```text
grok plugin marketplace add 88plug/claude-code-plugins
grok plugin install trigger-my-training@88plug --trust
```
The hard gate and the `ground-first` skill work immediately. Optional status-line
badge (`⏚ TMT:armed` / `:grounded`):
```bash
bash /path/to/trigger-my-training/install.sh
```
Full install and config: [docs](https://88plug.github.io/trigger-my-training/).
## What it does
Ask an AI coding agent to *"deploy a VM to Proxmox"* and it tends to barrel into
`qm create` — stale defaults, skipped prerequisites. This plugin intercepts
that pattern with agent guardrails and live verification:
1. **Recognize** (`ground-first` skill, model-elected) — the agent judges, from
its *own understanding*, whether a request is complex/irreversible enough to
ground — in any domain, not a keyword list. (Measured: precision 1.0 / recall
0.97 across infra + diverse domains, vs a keyword classifier's 0.28 recall
off-infra.)
2. **Reconstruct** — split what you **KNOW** from what you're **ASSUMING**, emit a
**Grounding Brief** (decision points, silent failure modes, unknowns tagged
**PROBE** / **ASK** / **ASSUME**), and verify your three riskiest assumptions.
3. **Enforce** (`PreToolUse` hook, self-arming) — a destructive action is
**denied** until grounding is recorded with `tmt-ground commit`. No detector
arms it; read-only probes and ordinary file edits are never blocked.
> [!NOTE]
> The *trigger* is the model's judgment (no hand-authored keyword list — that
> would be the static domain knowledge this plugin replaces). The *gate* stays
> deterministic: a safety floor must not depend on the model it is gating.
| Component | Surface | Role |
| --- | --- | --- |
| `ground-first` | skill (+ 6 reference packs) | Soft trigger — model-elected, keyword-free; holds the Grounding Brief procedure |
| `tmt_enforce.py` | PreToolUse hook | Hard deny on the irreversible step |
| `tmt_reconcile.py` | PostToolUseFailure hook | On tool failure, reconcile the falsified assumption (predict-then-check) |
| `tmt_log.py` / `tmt_session.py` | PostToolUse / SessionStart | Record probes / prune stale state |
| `tmt-ground` | bin CLI | Release the gate after probing |
| `grounding-investigator` | agent | Isolated live-probing pass for CRITICAL tasks |
| `tmt_statusline.sh` | status line | `⏚ TMT:armed` / `⏚ TMT:grounded` badge |
| `/status` `/ground` `/reset` `/brief` `/explain` `/doctor` | commands | Inspect / force / disarm / brief / explain / health-check |
The advisory/hard split is deliberate: the nudge raises the odds the agent
grounds; the `PreToolUse` deny is the lever that actually holds.
### Gate modes and self-arm
| `gate_mode` | Hard gate | Typical use |
| --- | --- | --- |
| `full` (default) | **on** | production — brief + gate |
| `gate` | **on** | gate only (eval / ablation) |
| `brief` | off | soft reflex only |
| `stale` | off | eval ablation arm |
| `off` | off | hard path disabled |
| option | default | effect |
| --- | --- | --- |
| `hard_gate` | `true` | `false` makes the gate advisory-only (keeps the skill, drops the block) |
> [!WARNING]
> Defaults are the hard deny. Self-arm is **lazy and detector-free**: the first
> mutating tool call is denied and marks the session `required`. Read-only probes
> and local file edits always pass. `tmt-ground commit` refuses without probe
> evidence (use `--force` only when probing is impossible). Details:
> [Architecture — self-arm](https://88plug.github.io/trigger-my-training/architecture/)
> · [Configuration](https://88plug.github.io/trigger-my-training/userconfig-design/).
Plugins can't register a main `statusLine`, so the badge ships in `bin/` and is
wired via `install.sh` — see [docs](https://88plug.github.io/trigger-my-training/).
## Does it work?
Built falsification-first. Full ledger: [`EXPERIMENTS.md`](EXPERIMENTS.md).
- **Detector (Exp 2, deterministic):** clean separation of operational vs
edit-intent on a 28-task labelled corpus — precision/recall **1.0**, **0**
false positives (in-sample; real-world calibration is known debt).
- **Hard gate (Exp 3, unit):** blocks the mutation, allows probes + local
edits, releases after grounding — **7/7**.
- **Landmine-catch (Exp 1, A/B + ablation, powered to 12 domains):** replicated
on `claude-haiku-4-5` —
| arm | catch-rate | vs baseline |
| --- | --- | --- |
| no plugin | 0.181 | — |
| **compact Pre-Mortem Brief** | **0.386** | **~2.1×** |
| + Staleness Axiom alone | 0.156 | worse (axiom alone does nothing) |
| enriched (4 composed inventions) | 0.258 | worse than the plain brief |
The campaign **falsified its own maximalist hypothesis**: a one-line "your
training is stale" axiom does nothing, and *composing* more proven cognitive
scaffolds (Tetlock calibration tags + Deming predict-then-check + Deutsch
hard-to-vary) **regressed** the gain. The active ingredient is one 3-line
structural trigger — Klein's pre-mortem + Popper's "enumerate how it breaks."
**Adding cognitive mass crowds it out.** That result survived replication
across 12 domains; an exciting single-run "win" for the enriched slate did
not (Twyman's law). Honest accounting of "10×": the replicated grounding
number is **~2.1×**; the only *literal* ≥10× is the poka-yoke gate taking
irreversible-action interception from ~0 to ~1.0. Full ledger:
[`EXPERIMENTS.md`](EXPERIMENTS.md), invention slate: [`INVENTIONS.md`](INVENTIONS.md).
```bash
bash evals/run.sh # all three experiments
python3 evals/detector_eval.py # free deterministic detector eval
```
## What this is not
Three independent refuters killed the maximalist pitch; the design reflects it:
- It does **not** claim the model "already knows" the domain — training is
stale; that is the whole point.
- It does **not** replace human-authored skills for must-be-exact execution —
it grounds the *understanding* layer.
- The reasoning mechanism is **not** novel (step-back / generated-knowledge /
preflight prior art) — the contribution is the packaging and the
harness-enforced gate.
## Layout
```
.claude-plugin/ plugin.json, marketplace.json
bin/ enforcer, probe-log, tmt-ground state machine, lib
hooks/ hooks.json (SessionStart, PreToolUse, PostToolUse, PostToolUseFailure)
skills/ground-first/ SKILL.md + reference packs
agents/ grounding-investigator.md
commands/ status, ground, reset, brief, explain, doctor
evals/ detector_eval.py, gate_unit_test.sh, tasks.jsonl
docs/ MkDocs Material site (this repo's GitHub Pages)
EXPERIMENTS.md the falsification ledger
```
## Docs
- Site: [https://88plug.github.io/trigger-my-training/](https://88plug.github.io/trigger-my-training/)
- [Architecture](docs/architecture.md) — detect→brief→gate→release + self-arm
- [Configuration](docs/userconfig-design.md) — `gate_mode` / `hard_gate`
- [Research](docs/research.md) — headline results + reproduce commands
## Development
Local checkout (no marketplace):
```bash
claude --plugin-dir /path/to/trigger-my-training
```
```bash
bash tests/run.sh # unit tests
bash evals/run.sh # the experiments
claude plugin validate . # manifest check
mkdocs build --strict # docs site
```
See [`CONTRIBUTING.md`](CONTRIBUTING.md) for the bin/hook architecture,
[`CHANGELOG.md`](CHANGELOG.md), and [`SECURITY.md`](SECURITY.md).
## License
[FSL-1.1-ALv2](LICENSE) © 2026 [88plug](https://github.com/88plug) —
Functional Source License; converts to Apache 2.0 two years after each release.