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 HelixDevelopment/code --skill action-prefix-systemgit clone --depth 1 https://github.com/HelixDevelopment/codeWrote 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/helixdevelopment/code/action-prefix-system)<a href="https://agentmods.dev/skills/helixdevelopment/code/action-prefix-system"><img src="https://agentmods.dev/badge/skills/helixdevelopment/code/action-prefix-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/helixdevelopment/code/action-prefix-system"><img src="https://agentmods.dev/badge/skills/helixdevelopment/code/action-prefix-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 | $0.00157 | $0.01962 |
| Opus 5.5 | $0.00063 | $0.00785 |
| Sonnet 5.5 | $0.00031 | $0.00392 |
| Haiku 4.5 | $0.00016 | $0.00196 |
Grade A, and why
Action Prefix System 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 2d 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 — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Action Prefix System (§11.4.140)
A prompt whose first non-blank line starts with a registered action token is
not an ordinary prompt: the token is replaced by that action's registered
expansion text, the action's rules apply, and the remainder of the prompt is
the actual task, executed under the expanded instruction.
The registry — constitution/actions/registry.yaml (override with
$HELIX_ACTION_REGISTRY) — is the single source of truth. It is DATA:
adding an action is a registry row, never a code change.
The six equivalent forms
All six resolve to the SAME action, the SAME expansion, the SAME execution:
| # | Form | Example |
|---|---|---|
| 1 | ACTION :: <rest> |
BACKGROUND :: refactor the parser |
| 2 | PREFIX::ACTION :: <rest> |
DEFAULT::BACKGROUND :: refactor the parser |
| 3 | /ACTION <rest> |
/background refactor the parser |
| 4 | /PREFIX::ACTION <rest> |
/DEFAULT::BACKGROUND refactor the parser |
| 5 | ACTION ---> <rest> |
BACKGROUND ---> refactor the parser |
| 6 | ACTION: <rest> — single colon |
BUG: subtitles render one frame late |
PREFIX is a namespace; the reserved default namespace is DEFAULT. An
action runs with or without it.
Form 6 is the one most people type naturally (BUG: ..., TASK: ...) — a
colon and exactly one space, no space before the colon.
Form 6 is REGISTERED-ACTION-ONLY by construction
Unlike the other five, an unknown single-colon token is a silent no-op — it
never asks for clarification and never resolves to a sub-system
(single_colon_registered_only: true). This is deliberate: ordinary prose like
TODO: fix this later, NOTE: see below, or WARNING: slow path MUST NOT
trigger a directive or a clarifying question. Only a token that is already
registered activates in form 6.
Grammar (all of these hold)
- Anchored to the first non-blank line only. A token mid-prose NEVER matches.
- UPPERCASE only —
[A-Z][A-Z0-9_]*. Lowercase never matches. (File and directory names stay lowercase snake_case per §11.4.29 — the token is upper.) - Namespace separator
::carries NO surrounding spaces (PREFIX::ACTION), which is what distinguishes it from the body separator" :: "(exactly one ASCII space each side). That spacing rule is why C++Foo::Bar, YAMLkey: value, and URLs never false-trigger. - Body separators:
" :: "(forms 1/2)," ---> "(form 5), one space (forms 3/4), and": "— colon + exactly one space, no space before the colon — (form 6). - Stacked prefixes apply outer-to-inner, left-to-right:
A :: B :: restexpandsA, re-scans, expandsB, and the residual is the task. - Escape with a leading
\—\BACKGROUND :: x,\BACKGROUND ---> x,\/BACKGROUND xare literal: strip the backslash, NO expansion. Use this to discuss an action without invoking it.
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.
- 2d ago Changed · +81 lines · +132 tokens per session 9fed26b0b40f
- 1mo ago First seen · 54 lines · 25 tokens per session scan A 9f1b5fe7b11e
Action Prefix System is a skill published in the GitHub repository HelixDevelopment/code (2 stars, last pushed 2d ago), licensed MIT. It adds 157 tokens to every session and 1,962 once invoked, about $0.0006 per session on Opus 5.5. 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.
Other skills, from other repositories
hqe
Comprehensive codebase health auditing, remediation, and verification skill based on the canonical HQE Protocol v5.0.0.
kirocrew-app-dev
Build, package, and publish Kiro Crew external apps. Covers app.json manifests, UI components, crons, skills, lifecycle hooks, development mode, registry publishing, and common pitfalls.
goal-conductor
Own a goal too large for one session ('clear the flaky-test backlog', 'push N PRs green') end to end via the work ledger — decompose into items, one session per item, read worker status as data, verify claims with the acceptance evaluator, decide each round until done. Use when handed such a goal.
cost-efficiency-analyzer
Analyzes cost structure, cost efficiency, and expense management from P&L data. Use when the user asks about costs, expenses, COGS, operating expenses, cost ratios, cost control, spending efficiency, margin compression from cost side, or wants to understand where money is going. Also use for "are we spending too…
security-conductor
Operating procedure for the kirocrew-security-conductor agent. Use when a security conductor session is seeded, inspected or debugged. Run vulnerability discovery on one target as a supervised fleet — one auditor per attack surface, an independent verifier per finding, two human gates.
computer-use
Read and drive native desktop apps through the accessibility layer — list apps, snapshot a window as a numbered element tree, then click/type/scroll/drag by element index or screen coordinates. Use for a desktop app, not a web page. macOS + Windows; off unless enabled in Settings.