agentleFS
Nunchi-tradeNunchi-trade/agent-cli/skills/pulse/SKILL.md

pulse

Skill521 starsChanged 7 months agoMITView on GitHub

Detects assets with sudden capital inflow via OI/volume/funding proxy signals

21 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use ant, artistic and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • bergsideSkill

    ant

    Structured, enterprise-focused design system emphasizing clarity, consistency, and efficiency for data-dense web applications.

    Also in 3 projects that use this

  • bergsideSkill

    artistic

    High-contrast, expressive style with creative typography and bold color choices for visually striking interfaces.

    Also in 3 projects that use this

  • +10 more from bergside

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 18 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    All other signals require Radar confirmation (score > 170)

    line 54
  2. Distinctive:

    ALWAYS check baseline history — signals from assets with < 2 scans are unreliable

    line 55
  3. Distinctive:

    NEVER act on erratic assets (rank bouncing between scans)

    line 56

“Common” is measured: kept by at least 6 in 10 of the 18 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Pulse — Capital Inflow Detector

Identifies assets accelerating in capital concentration before they become crowded positions. Uses publicly available HL market data as proxy signals for institutional flow detection.

Signal Types

SignalTriggerConfidence
IMMEDIATE_MOVEROI +15% AND volume 5x surge100
VOLUME_SURGERecent 4h volume > 3x average70
OI_BREAKOUTOI jumps 8%+ above baseline60
FUNDING_FLIPFunding rate reversal or 50%+ acceleration50

Direction Classification

Majority vote across available signals:

  • Funding rate sign -> directional bias
  • Price breakout direction
  • Volume surge + price momentum

Quality Filters

  1. Erratic detection (rank bouncing -> filtered)
  2. Minimum 24h volume ($500K default)
  3. Minimum scan history for baseline (2 scans)

Usage

hl pulse once              # Single scan
hl pulse run --tick 60     # Continuous (60s intervals)
hl pulse once --json       # JSON output
hl pulse once --mock       # Mock data
hl pulse status            # Last scan results
hl pulse presets           # List presets

Agent Mandate

You are the Pulse capital inflow detector. Your job is to catch capital inflow signals BEFORE the crowd. You detect timing — Radar detects setups. Together they form the APEX entry pipeline.

RULES:

  • IMMEDIATE_MOVER is the only signal strong enough for standalone entry
  • All other signals require Radar confirmation (score > 170)
  • ALWAYS check baseline history — signals from assets with < 2 scans are unreliable
  • NEVER act on erratic assets (rank bouncing between scans)
  • Report direction with every signal — directionless signals are useless

Decision Rules

Signal TypeConfidenceStandalone Entry?Action
IMMEDIATE_MOVER100YESEnter immediately — rare, high-conviction
VOLUME_SURGE70NOCheck Radar score. Enter if > 170
OI_BREAKOUT60NOCheck Radar score. Enter if > 200
FUNDING_FLIP50NOInformational only — do not enter on funding alone
Multiple signals same assetVariesYES if combined > 150Compound conviction — enter with caution
Baseline StateAction
< 2 scans in historyIgnore all signals — insufficient baseline
Erratic rank changesIgnore — likely data noise or manipulation
Consistent accelerationHigh confidence — act on signal
Signal appeared then disappearedStale — do not enter

Anti-Patterns

  • Acting on FUNDING_FLIP alone: Funding flips are noisy and low-conviction. Only useful as confirming signal.
  • Entering on VOLUME_SURGE without Radar: Volume surges occur on news, liquidations, and wash trading. 30% are false positives without Radar confirmation.
  • Ignoring the erratic filter: Assets with bouncing rankings are being manipulated or have unstable liquidity. The erratic filter exists for a reason.
  • Running too frequently: Pulse needs time between scans to build baselines. Running every 10s wastes API calls without improving signal quality. Use 60s minimum.

Error Recovery

ErrorCauseFix
No signals detectedQuiet marketNormal — no action needed
Baseline too shortFirst 2 scansWait for more scan history to accumulate
API timeout on candle fetchRate limitIncrease tick interval or reduce parallel fetches
Stale signal (> 5 min old)Delayed processingRe-scan before acting — signal may have decayed

Composition

Pulse is a sub-component of APEX (runs every tick). Pairs with Radar — Radar identifies high-quality setups, Pulse detects the optimal entry timing via capital flow signals. When used standalone, always cross-check with hl radar once.

Cron Template

# Standalone Pulse scan every 60s during trading hours
* 8-20 * * 1-5 cd ~/agent-cli && hl pulse once --json >> data/pulse/signals.jsonl 2>&1

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.