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/ar6420/hail_hydra/statsgit clone --depth 1 https://github.com/AR6420/Hail_HydraWrote 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/ar6420/hail_hydra/stats)<a href="https://agentmods.dev/commands/ar6420/hail_hydra/stats"><img src="https://agentmods.dev/badge/commands/ar6420/hail_hydra/stats.svg" alt="Measured on agentmods" 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 | $0.00029 | $0.02171 |
| Opus 5 | $0.00015 | $0.01086 |
| Sonnet 5 | $0.00006 | $0.00434 |
| Haiku 4.5 | $0.00003 | $0.00217 |
Grade A, and why
stats 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 4d 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 — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hydra Stats — Real Token Tracking
Read the active Claude Code session log and compute actual token usage and savings. NO AI estimation — pure JSONL parsing.
Math + JSONL parsing live in the shared helper at
~/.claude/hooks/hydra-token-math.js. Statusline and /hydra:stats both
call it so numbers stay consistent.
Pricing (per 1M tokens, verified 2026-08 for Claude 4.x / 5.x)
| Tier | Input | Output | Cache read |
|---|---|---|---|
| Haiku | $1 | $5 | 10% of input |
| Sonnet | $3 | $15 | 10% of input |
| Opus | $5 | $25 | 10% of input |
| Fable | $10 | $50 | 10% of input |
Prices are keyed by tier, not by model ID — Opus 4.6 through Opus 5 all bill at
$5/$25, so point releases need no change. Edit the PRICING map in
hydra-token-math.js only if Anthropic publishes new prices or ships a new tier.
Fable sits above the Opus baseline, so Fable turns are not counted as delegations and they reduce (never increase) reported savings.
Run
# Strikethrough capability detection — env-heuristic only.
USE_STRIKETHROUGH=0
[ "$TERM_PROGRAM" = "Apple_Terminal" ] && USE_STRIKETHROUGH=1
[ "$TERM_PROGRAM" = "iTerm.app" ] && USE_STRIKETHROUGH=1
[ "$TERM_PROGRAM" = "vscode" ] && USE_STRIKETHROUGH=1
[ -n "$KITTY_WINDOW_ID" ] && USE_STRIKETHROUGH=1
[ "$TERM" = "alacritty" ] && USE_STRIKETHROUGH=1
[ -n "$WEZTERM_PANE" ] && USE_STRIKETHROUGH=1
[ -n "$WT_SESSION" ] && USE_STRIKETHROUGH=1
# Known-incompatible terminals (force fallback, overrides green-list)
[ -n "$MSYSTEM" ] && USE_STRIKETHROUGH=0
[ -n "$CYGWIN" ] && USE_STRIKETHROUGH=0
echo "$TERM" | grep -q "cygwin" && USE_STRIKETHROUGH=0
# User override
[ "$HYDRA_STRIKETHROUGH" = "0" ] && USE_STRIKETHROUGH=0
[ "$HYDRA_STRIKETHROUGH" = "1" ] && USE_STRIKETHROUGH=1
HYDRA_USE_STRIKETHROUGH="$USE_STRIKETHROUGH" node -e "
const path = require('path');
const os = require('os');
const helperPath = path.join(os.homedir(), '.claude', 'hooks', 'hydra-token-math.js');
let tokenMath;
try {
tokenMath = require(helperPath);
} catch (e) {
console.log('hydra-token-math.js not installed at ' + helperPath);
console.log('Run: hail-hydra-cc to (re)install Hydra hooks.');
process.exit(0);
}
const summary = tokenMath.computeSummary();
if (!summary.available) {
console.log('No session data for this project yet.');
process.exit(0);
}
const useStrike = process.env.HYDRA_USE_STRIKETHROUGH === '1';
const STRIKE = useStrike ? '\x1b[9m' : '';
const STRIKE_OFF = useStrike ? '\x1b[29m' : '';
const GREEN = '\x1b[32m';
const BOLD = '\x1b[1m';
const DIM = '\x1b[2m';
const RESET = '\x1b[0m';
function fmt(n) {
if (n >= 1_000_000) return (n / 1_000_000).toFixed(2) + 'M';
if (n >= 1_000) return (n / 1_000).toFixed(1) + 'k';
return n.toString();
}
const { stats, totalTurns, haikuCost, sonnetCost, opusCost, fableCost,
actualCost, hypotheticalCost, savedUSD, savedPct,
delegatedTurns, delegationRate, sessionFile, unknownModels } = summary;
const bar = '━'.repeat(40);
// No-delegation guidance branch — when no Hydra subagents dispatched OR savings below indicator threshold
if (delegatedTurns === 0 || savedUSD < 0.01) {
console.log('');
console.log('🐉 Hydra Stats');
console.log(bar);
console.log('Session: ' + path.basename(sessionFile));
console.log('Turns: ' + totalTurns);
console.log(bar);
console.log('');
console.log('🟣 Opus (' + stats.opus.turns + ' turns): ' + fmt(stats.opus.input + stats.opus.cache_create) + ' in / ' + fmt(stats.opus.output) + ' out → \$' + opusCost.toFixed(3));
if (stats.fable.turns > 0) {
console.log('🔴 Fable (' + stats.fable.turns + ' turns): ' + fmt(stats.fable.input + stats.fable.cache_create) + ' in / ' + fmt(stats.fable.output) + ' out → \$' + fableCost.toFixed(3));
}
console.log(bar);
console.log('');
console.log('No Hydra subagent dispatches recorded in this session.');
console.log('');
console.log('Hydra works best when invoked explicitly. Try:');
console.log(' /hydra:guard — security scan on Haiku');
console.log(' /hydra:preflight — environment validation');
console.log(' /hydra:map — codebase dependency map');
console.log('');
console.log('Or include \"use hydra\" in prompts that involve multi-file');
console.log('exploration, codebase analysis, or routine verification.');
console.log(bar);
if (unknownModels && unknownModels.size > 0) {
console.log('');
console.log('⚠️ Unknown models (not counted): ' + Array.from(unknownModels).join(', '));
}
process.exit(0);
}
console.log('');
console.log('🐉 Hydra Stats');
console.log(bar);
console.log('Session: ' + path.basename(sessionFile));
console.log('Turns: ' + totalTurns);
console.log(bar);
console.log('');
console.log('🟢 Haiku (' + stats.haiku.turns + ' turns): ' + fmt(stats.haiku.input + stats.haiku.cache_create) + ' in / ' + fmt(stats.haiku.output) + ' out → \$' + haikuCost.toFixed(3));
console.log('🔵 Sonnet (' + stats.sonnet.turns + ' turns): ' + fmt(stats.sonnet.input + stats.sonnet.cache_create) + ' in / ' + fmt(stats.sonnet.output) + ' out → \$' + sonnetCost.toFixed(3));
console.log('🟣 Opus (' + stats.opus.turns + ' turns): ' + fmt(stats.opus.input + stats.opus.cache_create) + ' in / ' + fmt(stats.opus.output) + ' out → \$' + opusCost.toFixed(3));
if (stats.fable.turns > 0) {
console.log('🔴 Fable (' + stats.fable.turns + ' turns): ' + fmt(stats.fable.input + stats.fable.cache_create) + ' in / ' + fmt(stats.fable.output) + ' out → \$' + fableCost.toFixed(3));
}
console.log(bar);
console.log('');
console.log('Delegation rate: ' + delegationRate.toFixed(1) + '% (' + delegatedTurns + '/' + totalTurns + ' turns)');
if (useStrike) {
console.log('Was: ' + DIM + STRIKE + '\$' + hypotheticalCost.toFixed(3) + STRIKE_OFF + RESET);
console.log('Now: ' + BOLD + GREEN + '\$' + actualCost.toFixed(3) + RESET);
} else {
console.log('Actual cost: \$' + actualCost.toFixed(3));
console.log('All-Opus baseline: \$' + hypotheticalCost.toFixed(3));
}
console.log(bar);
console.log('💰 ' + GREEN + 'Saved: \$' + savedUSD.toFixed(3) + ' (' + savedPct.toFixed(1) + '%)' + RESET);
console.log(bar);
console.log('');
console.log('Reads Claude Code session JSONL directly. No AI estimation.');
if (unknownModels && unknownModels.size > 0) {
console.log('');
console.log('⚠️ Unknown models (not counted): ' + Array.from(unknownModels).join(', '));
console.log(' Update PRICING map in ~/.claude/hooks/hydra-token-math.js');
}
"
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.
- 4d ago First seen · 169 lines · 29 tokens per session scan A 88ec1e4ba33e
stats is a command published in the GitHub repository AR6420/Hail_Hydra (48 stars, last pushed 23d ago), licensed MIT. It adds 29 tokens to every session and 2,171 once invoked, about $0.0001 per session on Opus 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 2026-08-30.
Other commands, from other repositories
install-context
Install OpenAgents Control context files.
health
Run documentation health checks (freshness, links, drift, cross-doc consistency).
tidy
Consistency check for non-code repos with INDEX.md hierarchy. Verifies INDEX.md accuracy, MEMORY.md references, orphaned files, stale dates, and WAITING markers.
batch-execute
Execute a batch of ops plans continuously without user input, storing any issues and user questions in a .md file to be processed with the user afterwards. Do NOT invoke this yourself! Should only be invoked by the user.
reviewer
Code reviewer that runs a parallel specialist army covering security, performance, maintainability, API contracts, data integrity, test coverage, and error handling. Trigger on code review, review, PR review, pull request, or review army.
konzept-update
This command updates an existing rough concept based on new findings, clarified open questions, or changed requirements. Each revision is versioned and timestamped so that the evolution of the concept remains traceable.