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 agentmods add commands/flyfission/nuclear-grade-context-engineering/ng-intentgit clone --depth 1 https://github.com/FlyFission/nuclear-grade-context-engineeringWhat 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 | $0.00000 | $0.00622 |
| Opus 5 | $0.00000 | $0.00311 |
| Sonnet 5 | $0.00000 | $0.00124 |
| Haiku 4.5 | $0.00000 | $0.00062 |
Grade A, and why
ng-intent 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.
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ng-intent
Portable command prompt generated from skills/declaring-intent/SKILL.md. Edit the skill, then run python tools/ng.py gen-commands; do not edit this file by hand.
States what an agent intends to do and why before a critical or irreversible action, so a reviewer can challenge the thinking and not just the result. Use before deploys, migrations, public claims, or trust changes that deserve a stated intent, expected result, abort criteria, and backup. Do not use for routine reversible edits, and never treat the stated intent as proof the agent understood.
Use when
- Before a deploy, migration, data change, public claim, dependency or model swap, or release.
- When an action is hard to reverse, or its blast radius is more than the immediate file.
- When you want a reviewer's eyes on the reasoning and the abort criteria before execution.
- When standing authority lets an agent proceed unless told no, and the team needs the "no" window.
Do not use when
- The action is a routine, reversible edit with obvious proof and no trust boundary.
- An incident is live and the next move is stabilization, not a fresh proposal.
- Someone wants the declaration treated as a guarantee or certification.
Inputs
- The intended action and the exact target.
- The reasoning and the evidence that says the preconditions are met.
- The expected result, and the signals that would mean it went wrong.
- The abort and rollback criteria, the decision rights, and who is backing up the action.
Prompt text
Declare intent before this action the Nuclear-grade way.
Inputs:
- intended action and target:
- reasoning / evidence preconditions are met:
- expected result:
- abort criteria (numbers where possible):
- rollback (and is it verified?):
- who may stop it, by when / backup watcher:
Return:
- "I intend to <action> on <target> because <evidence/reasoning>"
- the expected result and the precise signals that would falsify it
- the abort criteria and the verified rollback
- the decision rights and the backup
- after acting: the actual result compared to the expected result, and any gap
State what would prove this wrong, not just what success looks like. Treat the stated intent as a proposal to review, not proof the agent understood. Do not imply certification or formal assurance.
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.
- 3d ago First seen · 59 lines · 0 tokens per session scan A 3a7fdb679fa7
ng-intent is a command published in the GitHub repository FlyFission/nuclear-grade-context-engineering (33 stars, last pushed 24d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 622 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 2026-08-30.
Other commands, from other repositories
coder-eval-implement-plan
Implement an approved codereval plan phase by phase with risk-scaled per-phase review, then a final code review.
capture-feedback
Quick feedback capture with structured signals.
auto-run
PitWay: Manage auto-run authorization for automatic task continuation.
task-integrate
PitWay: Apply a dispatched task's worktree commit to the main tree.
milestone-list
PitWay: List every milestone with a concise one-line status.
fix
Universal debugging and fix application with semantic code analysis.