aget-create-project
Create research projects with context-aware scaffolding. Reads AGET identity, searches prior work, pre-populates Due Diligence. Use when starting any formal AGET project.
npx skills add majiayu000/claude-skill-registry --skill aget-create-project --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Creates a new AGET project with research-informed scaffolding per PROJECT_PLAN_AGET_CREATE_PROJECT_V2.md.
How it works
- Determines mode from INPUT: interactive prompts if blank, explicit type/name if provided, or infers from topic; supports evidence-rich mode when input contains quotes like L-doc references.
- Performs Step 0 Scope-Fit Validation by reading .aget/identity.json and governance/SCOPE_BOUNDARIES.md to ensure alignment, warning only if misaligned and allowing override.
- Parses INPUT to identify project type (research or poc) and project name; in headless mode requires both and errors if missing.
- Checks for existing projects in planning/ and warns if a conflict is found.
- Research Phase 3.x gathers context:
- 3.1 Reads AGET Identity from .aget/identity.json (north_star, domain, archetype)
- 3.2 Searches relevant L-docs with grep -l "<topic_keywords>" .aget/evolution/*.md and extracts L-doc ID/title, key lesson, application
- 3.3 Searches prior sessions via grep -l "<topic_keywords>" sessions/*.md
- 3.4 Searches similar projects via grep -l "<topic_keywords>" planning/PROJECT_PLAN_*.md and WARNs if duplicates
- 3.5 Extracts vocabulary terms via grep -B2 -A5 "<topic_keywords>" specs/CLI_VOCABULARY.md
- Step 3.6 reads governing spec if applicable and may create a conformance baseline (not detailed here in output)
- Step 4 selects/validates templates, including a check for Governing Spec and conformance of template; stops on non-conformance.
- Step 5 scaffolds the plan by populating Phase -1 with L-doc results, includes traceability data and identity reference.
- Step 6 writes output to planning/PROJECT_PLAN_<NAME>.md for a research project or planning/poc-proposals/POC-XXX-<name>.md for a poc.
- Step 7 reports a summary including identity, related docs/sessions, and pre-populated content.
- Step 7.5 performs self-verification to ensure required sections exist and are consistent with the template if a governing spec is present.
- Step 8 checks gate ordering to ensure spec-first ordering (Gate -1) precedes implementation gates, warning if misordered.
When to use it
Use when starting any formal AGET project to generate context-aware scaffolding and pre-populated Due Diligence based on identity and evidence.
What it can touch
- Reads: .aget/identity.json
- Accesses: governance/SCOPE_BOUNDARIES.md
- Searches: .aget/evolution/.md, sessions/.md, planning/PROJECT_PLAN_*.md, specs/CLI_VOCABULARY.md
- Writes: planning/PROJECT_PLAN_<NAME>.md or planning/poc-proposals/POC-XXX-<name>.md
- Executes: shell-like commands listed in the SPEC (e.g., grep, cat, test)
Caveats
- DOES read governing specs and may stop if template violates SHALL requirements (L644).
- Requires governance and identity files to be present; headless mode without type/name is error.
- Does not overwrite existing projects without explicit confirmation.
- Template verification depends on presence of a Governing Spec field; absence yields a WARN but proceeds.
# /aget-create-project Create a new AGET project with research-informed scaffolding per PROJECT_PLAN_AGET_CREATE_PROJECT_V2.md. ## Input $ARGUMENTS ## Mode Detection | Input Pattern | Mode | Behavior | |---------------|------|----------| | Empty or blank | **Interactive** | Prompt for type and name | | `<type> <name>` | **Explicit** | Create project of specified type | | `<topic>` only | **Inference** | Infer type from topic and AGET domain | | Input contains evidence statements (interview data, gap analysis, L-doc citations, cross-reference synthesis) | **Evidence-rich** | Abbreviated research phase — use provided evidence, skip redundant searches | ### Evidence-Rich Mode **Detection**: Input contains 2+ of: specific L-doc references (L###), quantitative data (percentages, counts, session numbers), named sources (supervisor interviews, audit results), or cross-reference synthesis ("both supervisors agree..."). **Behavior**: Research Phase (Step 3) still executes per C-CP-003, but in abbreviated form: - Step 3.1 (identity): **Runs** — always needed for project context - Step 3.2 (L-docs): **Abbreviated** — search only L-docs explicitly cited in input, don't broad-search by topic ke
- Input
- Mode Detection
- Evidence-Rich Mode
- Supported Project Types
- Project Creation Process
- Step 0: Scope-Fit Validation (#295)
- Step 1: Input Analysis
- Step 2: Conflict Check
- Step 3: Research Phase (KEY DIFFERENTIATOR)
- Step 3.6: Spec-First Conformance Baseline (L616, L617, #313)
- Step 3.7: ADR-008 Precondition Check (#313)
- Step 3.8: Artifact Type Fitness Assessment (#315)
- Step 4: Template Selection and Scope Estimation (#391)
- Step 4.5: Template Spec-Conformance Verification
ls planning/ | grep -qi "<name>" cat .aget/identity.json grep -l "<topic_keywords>" .aget/evolution/*.md grep -l "<topic_keywords>" sessions/*.md grep -l "<topic_keywords>" planning/PROJECT_PLAN_*.md grep -B2 -A5 "<topic_keywords>" specs/CLI_VOCABULARY.md test -f templates/poc/RESEARCH_PROJECT_PLAN.template.md
What does the aget-create-project skill do?
Create research projects with context-aware scaffolding. Reads AGET identity, searches prior work, pre-populates Due Diligence. Use when starting any formal AGET project.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill aget-create-project --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.