agentleFS
tjdwls101010tjdwls101010/Harness-of-Minervini/.claude/skills/ticker-analysis/SKILL.md

ticker-analysis

Skill1 starsChanged 3 months agoMITView on GitHub

Analyze one or a few named US-listed stocks for a prospective buy, entry timing, setup diagnosis, active-position HOLD or SELL evidence, re-entry, earnings risk, chart condition, or head-to-head comparison. Use whenever the user asks what to do with a specific US-listed ticker, even if Minervini or SEPA is not named. Do not use for market-wide sector screening or leader discovery; use market-scan. Crypto, non-US listings, completed-trade grading, account-level position sizing, and capital allocation advice are outside scope.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use doctrine list --context <ctx> to discover the claims an analysis context rests on, then doctrine show <id> to read one.

    line 11
  2. Distinctive:

    Treat a virtual or fixed-evidence prompt as a closed world: use only its supplied evidence, never an unrelated fixture, live ticker, or web …

    line 14
  3. Distinctive:

    If eligible, examine weekly structure before daily timing.

    line 21
  4. Distinctive:

    Use the risk reducer for the final prospective state.

    line 26
  5. Distinctive:

    Use the user's actual entry date, entry price, hard stop, and structural invalidation.

    line 30
  6. Distinctive:

    Use web search only for current catalyst, company, earnings-event, or industry narrative, and only after the deterministic evidence is in.

    line 39
  7. Distinctive:

    If the evidence cannot support a decision, say exactly what remains unresolved.

    line 47

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “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

Ticker analysis

Orient through the interface

  • Use doctrine list --context <ctx> to discover the claims an analysis context rests on, then doctrine show <id> to read one.
  • Work from the repository root; the root constitution holds the bootstrap and capability-catalog commands. Before invoking a capability for the first time, read only its describe output or that leaf command's --help.
  • Decide whether the request is prospective, active-position, re-entry, or comparison. Collect only evidence that can change that decision.
  • Treat a virtual or fixed-evidence prompt as a closed world: use only its supplied evidence, never an unrelated fixture, live ticker, or web number. If a Primary Base duration, depth, or all-time-high trigger is not supplied, keep it missing; do not infer it.

Prospective entry

  • Qualify first. Do not research a story, inspect valuation, or deepen the setup before the low-cost technical gate earns it.
  • On a known gate failure, stop and name the exact failed evidence. On an incomplete one, resolve only the named gap -- a chart for needs_chart -- and never infer a pass.
  • Declaring a plane's state is an assertion, and only a gate licenses one. A band or marker below its range is review evidence: typing it into --fundamentals-state or --setup-state turns a measurement that reports into a verdict that decides, and the reducer takes it because a caller's word is always allowed to be more cautious. Where no gate failed and the other planes are unmeasured, the ticker is INCOMPLETE -- a route you can rule out on its own terms is not the ticker's verdict.
  • If eligible, examine weekly structure before daily timing. Require separate price geometry, contracting supply, a completed pivot or VCP-anchored cheat, and precise invalidation.
  • A named VCP without supply evidence is incomplete. doctrine show tactic.<name> lists what a [TL-EARLY] tactic requires; each condition is a chart reading you declare, and one you could not settle is declared as such rather than left out.
  • Evaluate SEC filed-as-of fundamentals, accounting integrity, dilution, growth quality, and leadership category, then read the growth numbers under the leader category the filings support.
  • Compare the ticker with current same-industry peers.
  • Recheck the market evidence this setup needs. QQQ context alone cannot supply market alignment, and absent user trade traction remains a real gap.
  • Use the risk reducer for the final prospective state. Never call a ticker BUY-READY from qualification, setup, fundamentals, or RS alone.

Active position or re-entry

  • Use the user's actual entry date, entry price, hard stop, and structural invalidation. If anchors are missing, report INCOMPLETE and ask only for information that can change HOLD or SELL.
  • Audit every completed daily low from the stop's effective date through the analysis session. A recovered latest price cannot establish HOLD after an earlier breach; incomplete path coverage means INCOMPLETE. Use a partial-session breach only when the user explicitly requests a live stop check.
  • For HOLD, require both current completed-price evidence and a clear completed stop path. If the stop changed after entry, use its actual effective date rather than applying it retroactively.
  • Report the management_actions the envelope carries beside the verdict, each with the claim and measurement behind it. A HOLD with an untouched stop and deteriorating structure is a HOLD plus REVIEW, never a bare HOLD.
  • A re-entry is a new prospective decision. Re-run qualification, setup, market, and risk rather than treating a prior win or loss as permission.

Chart and narrative judgment

  • Render a chart only when qualitative ambiguity matters. Read weekly before daily, moving averages as zones, overhead supply, base maturity, price/volume character, and expected behavior.
  • Use web search only for current catalyst, company, earnings-event, or industry narrative, and only after the deterministic evidence is in.
  • Respect abnormal price weakness after apparently good news. Price can reveal deterioration before the public explanation.

Comparisons and synthesis

  • For a few named tickers, gate each independently before comparing. A failed ticker cannot win through relative scoring.
  • Compare only aligned evidence and dates: eligibility, setup quality, filed fundamentals, same-industry leadership, market fit, downside, and reward-to-risk.
  • Lead with BUY-READY, WAIT, AVOID, INCOMPLETE, HOLD, or SELL and evidence quality. Then separate known failures, missing evidence, qualitative chart judgment, entry or promotion conditions, and invalidation or exit conditions.
  • If the evidence cannot support a decision, say exactly what remains unresolved.

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.