# Goga Brainstorm Plan Verification > Verification of the assembled architecture plan against the DSL - **Type:** Skill - **Install:** `agentstack add skill-qarium-goga-goga-brainstorm-plan-verification` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [qarium](https://agentstack.voostack.com/s/qarium) - **Installs:** 0 - **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills) - **Latest version:** 0.1.0 - **License:** BSD-3-Clause - **Upstream author:** [qarium](https://github.com/qarium) - **Source:** https://github.com/qarium/goga/tree/1.1.x/goga/assets/skills/goga-brainstorm-plan-verification ## Install ```sh agentstack add skill-qarium-goga-goga-brainstorm-plan-verification ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # goga-brainstorm-plan-verification ## Identity You are responsible for verifying the created architecture plan and fixing the issues found. ## Context Use these skills for verification: - **`goga-cell`** and **`goga-cookbook`** — DSL validation (syntax, directives, design principles). - **`goga-codemanifest-base`** — base usages/annotations compliance. - **`goga-lang-disp`** — language correctness (naming, `location` file names). Use this artifact for its specific purpose: - **`[ARCHITECTURE_PLAN]`** (at `docs/arch/.md`) — use it as the **object of verification**: its implementation order, per-cell CODEMANIFESTs and `.usages/` files, dependency map, and verification checklist, against which the DSL checks are run, failures are fixed in place, and the report is produced. ## Workflow ### Phase 1. Run all verification checks Run every check in **Verification Checks** below. For each, record PASS/FAIL with evidence. ### Phase 2. Fix failures For each FAIL — fix the issue in the plan file and re-check. ### Phase 3. Produce the report Produce the `[VERIFICATION_REPORT]` (see Output Format). ## WAIT Present the final (fixed) plan and the `[VERIFICATION_REPORT]` to the user and obtain final confirmation before the pipeline concludes. - **Confirmed** → the pipeline is complete - **User requests changes** → address the feedback, re-verify, and present again ### Verification Checks 1. **Completeness** — every type, method, property from the approved solution is present in the plan 2. **DSL correctness** — all CODEMANIFESTs syntactically correct (keys, signatures, document structure) 3. **Inter-cell consistency** — Imports reference existing cells, types match 4. **Implementation order** — each cell created after all cells it depends on 5. **No placeholders** — no TBD, TODO, or incomplete descriptions in the plan's CODEMANIFEST files 6. **Usage of Imports.Types** — every imported type is used in the body (signatures, mutations, embeddings, annotations) 7. **Usage of Imports.Usages** — every imported practice is mentioned in at least one annotation 8. **Usage of Usages** — every practice declared in the `Usages` header is mentioned in at least one annotation 9. **Algorithms in annotations** — annotations for routines and methods describe the operation algorithm if achievable 10. **Annotation wording** — annotations do not contain technical implementation details 11. **Resolvability of references in annotations** — every backtick reference resolves (signature variable, imported/declared type, Usages/Imports practice) 12. **Location restrictions** — every `location` value is a file name with extension only, no directories, no escaping the current level 13. **Absence of cross-imports** — if cell A imports from cell B, cell B does not import from cell A 14. **Embedding from Imports** — every embedded type (via `->`) is available through `Imports` 15. **Mutations from available types** — base types in mutations (`Object::Target`) are available (imported or declared) 16. **Entity / Routine correctness** — types with `methods`/`properties` are Entities; types without are Routines and lack these sections 17. **Base usages from configuration** — each base practice is included in `Usages` of all CODEMANIFESTs and referenced in at least one annotation 18. **Base annotations from configuration** — base annotations are included in `Annotations` of all CODEMANIFESTs and contracts comply with them 19. **Language correctness** — type/method/property names comply with target-language conventions; `location` values have correct file names for the target language ## Output Format Fill every section. No empty sections. ```md # [VERIFICATION_REPORT] ## Check Results [Table: # | Check | PASS/FAIL | Evidence] ## Fixes Applied [Table: Issue | Fix | Re-check status] ## Unresolved Issues [List any unresolved DSL errors or failed checks. Empty if none] ## Final Status [VERIFIED / FAILED — with justification] ``` ## STOP if: - unresolved DSL errors remain - any verification gate fails after attempted fix ## Source & license This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [qarium](https://github.com/qarium) - **Source:** [qarium/goga](https://github.com/qarium/goga) - **License:** BSD-3-Clause Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** no - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/skill-qarium-goga-goga-brainstorm-plan-verification - Seller: https://agentstack.voostack.com/s/qarium - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.