cilipep/ai-agent-with-tslab/strategy-plan
- community
Checking that the SKILL.md file is still in the repository.
Use after strategy design and before TSLab graph authoring to turn a strategy spec into a concrete execution plan with documentation, indicator, risk, lifecycle, and decision gates.
Install
npx vskill@latest install cilipep/ai-agent-with-tslab/strategy-planSKILL.md
Strategy Plan
Use this phase after strategy-design and before strategy-authoring.
The goal is to convert a strategy specification into a concrete execution plan that the agent can follow without skipping risk, documentation, indicator checks, or run analysis.
Existing Project Rules
Strategy Flow is a public methodology plugin. It should remain neutral with respect to any one trader, author, or private workspace.
If the active workspace already contains its own strategy-development standards, templates, or indicator rules, preserve them when they are stronger or more specific than the generic Strategy Flow defaults.
Do not create a competing documentation standard inside the same project when the project already has:
- a strategy description template;
- an indicator description template;
- a risk document;
- an indicator registry;
- project-specific implementation rules for the TSLab visual editor.
Required Plan Output
Produce a step-by-step execution plan that includes:
- create or locate the target script;
- map the source data;
- confirm stop contract and pane map before graph authoring;
- build the minimum meaningful trade path;
- add risk sizing and connect
Shares; - add initial stop and trailing/protective logic;
- clean the graph;
- run lifecycle proof;
- inspect metrics, messages, and logs;
- run final visual and pane audit after the last mutation;
- decide: repair, optimize, or finalize.
Every substantial mutation inside the plan must include a mini-loop:
- mutate;
- cleanup audit;
- lifecycle proof;
- decide whether the mutation is accepted or must be repaired immediately.
Required Documentation Checks
The plan must also include:
- confirm which starting document exists:
initial-strategy-description,source-strategy-description, orchange-description; - create or update the strategy documentation under the strategy folder;
- verify that the strategy logic is described in human language, not only as graph blocks;
- plan to produce a full
final-strategy-descriptionafter implementation, even when the task is only a modification of an existing strategy; - if a new custom indicator is created, create or update its own final indicator description;
- if a new custom indicator is created, update the indicator registry/list in the active project when that registry exists;
- if the active project already has a strategy-description or indicator-description standard, follow that standard instead of inventing a new one;
- preserve links between strategy documentation and indicator documentation.
- define whether the current goal is prototype, verified baseline, delivery-clean, or live-trading-ready.
Indicator Decision Gate
Before allowing a new indicator in the plan, explicitly check:
- can this be done with standard TSLab blocks;
- can this be done with formulas;
- does an existing documented project indicator already exist;
- is the new indicator a reusable primitive rather than a one-off strategy hack.
If the answer does not justify a new indicator, the plan must return to built-in blocks or formulas.
Plan Shape
Each plan step should say:
- what artifact is being changed;
- what proof will confirm the step;
- what blocker would force a repair loop instead of continuing.
The stop-contract step must explicitly say:
- source level for long and short stops;
- whether the formula may use
Close; - whether the stop is entry-anchored, dynamic protective, or monotonic trailing;
- whether any engineering stop is optional and how it is switched on or off.
The risk step must explicitly say:
- what capital base is used if risk is percent-based;
- whether that capital base comes from initial deposit, a constant, or another explicit source;
- whether risk parameters must be optimization-ready.
Exit Conditions
Do not allow the plan to jump directly from authoring to optimization.
The execution plan must explicitly pass through:
- risk complete;
- run analysis complete;
- graph cleanup complete;
- final visual and pane audit complete;
- documentation complete enough for finalization.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingStrategy-Flow/skills/strategy-plan/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/cilipep/ai-agent-with-tslab/strategy-plan)