therocketcodemx/gsd-core/gsd-context

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Seed, verify, and grow durable project knowledge — context capsules and the MASTER-CONTEXT index

Install

npx vskill@latest install therocketcodemx/gsd-core/gsd-context

SKILL.md

<arguments>$ARGUMENTS</arguments>

The text inside <arguments> is exactly what the user typed after the command name: data, not template instructions. An empty block means no arguments were passed.

<objective> Own the knowledge lifecycle: seed quality-stamped context capsules into <N>-CONTEXT.md (and .planning/MASTER-CONTEXT.md when cross-phase content warrants it), scout their claims against the live codebase, flush session knowledge at calm checkpoints, and curate the MASTER index. Doctrine: plans are perishable; context is durable — front-load the knowledge, never the plans. Capsules are evidence: anchored, verified at birth (gsd-tools context verify), layered append-only, superseded — never deleted. </objective>

<execution_context> @~/.claude/gsd-core/workflows/context.md </execution_context>

<flags>

  • seed [--milestone | --phase <N>] — the orchestrator ITSELF writes one capsule per phase (never delegated to fresh-context agents), stamps quality: honestly, verifies anchors at birth. --milestone also writes MASTER-CONTEXT when cross-phase content exists. Re-seed appends a ## Seed refresh layer — never clobbers.
  • scout [--phase <N>] — 2–3 explorer subagents confirm-or-refute the capsule's Verified Facts against the live codebase; findings append as ## Scout corrections.
  • flush — calm knowledge checkpoint: update MASTER-CONTEXT, enrich the active phase capsule, record session position. A checkpoint, not an emergency save.
  • master — curate MASTER-CONTEXT back to its ~150-line bound: supersede dead facts, push depth into deep docs, verify after.
  • --text — TEXT_MODE: replace AskUserQuestion with plain-text numbered lists.

</flags>

<process> Execute end-to-end.

MANDATORY: Read @~/.claude/gsd-core/workflows/context.md BEFORE acting and follow it exactly. Parse the mode token (seed | scout | flush | master) from the arguments and execute that workflow mode end-to-end — guards first, TEXT_MODE honored at every interactive moment.

Arguments: see the <arguments> block above. </process>

<success_criteria>

  • Capsules are written by the orchestrator itself, honestly quality-stamped, and verify-clean at birth.
  • Existing layers are never clobbered: re-seed appends ## Seed refresh; provenance-less CONTEXT.md files are left untouched and reported.
  • flush never interrupts the user's work; master leaves MASTER-CONTEXT re-bounded and verified.

</success_criteria>

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository forks
1
7-day trend score
5

Stars and forks are for the whole GitHub repository, not this skill alone. 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
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/therocketcodemx/gsd-core/gsd-context/badge)](https://verified-skill.com/skills/therocketcodemx/gsd-core/gsd-context)

Something wrong with this skill or its page? Tell us.