Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/griddynamics/rosettaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/griddynamics/rosetta/init-workspace-flow-verification)<a href="https://agentmods.dev/commands/griddynamics/rosetta/init-workspace-flow-verification"><img src="https://agentmods.dev/badge/commands/griddynamics/rosetta/init-workspace-flow-verification/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/griddynamics/rosetta/init-workspace-flow-verification"><img src="https://agentmods.dev/badge/commands/griddynamics/rosetta/init-workspace-flow-verification.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00016 | $0.01062 |
| Opus 5.5 | $0.00006 | $0.00425 |
| Sonnet 5.5 | $0.00003 | $0.00212 |
| Haiku 4.5 | $0.00002 | $0.00106 |
Grade A, and why
init-workspace-flow-verification scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<init_workspace_flow_verification>
<description_and_purpose> Without a final verification pass, incomplete or inconsistent documentation ships silently. Phase 9 runs a centralized checklist, ensures nothing was missed, and enforces new-chat requirement. </description_and_purpose>
<workflow_context>
- Phase 9 of 9 in init-workspace-flow (final phase)
- Prerequisite: Phases 1-8 complete
- Output: verification report, next steps, new-chat enforcement </workflow_context>
<phase_steps>
- Read state file and confirm prerequisites
- Execute verification checklist
- Suggest next steps
- Enforce new chat and mark COMPLETE </phase_steps>
<read_state step="9.1">
- Read
agents/init-workspace-flow-state.md - Confirm Phases 1-8 all marked complete
- Collect unresolved gaps from Phase 8 </read_state>
<execute_verification step="9.2" subagent="built-in" role="Workspace initialization auditor" subagent_recommended_model="claude-sonnet-5-5,gpt-6-luna-high">
Act as a senior workspace initialization auditor. This is the final phase of workspace initialization. Consolidates all init-phase outputs into a single completeness audit, runs catch-up for gaps, and revalidates assumptions.
<core_concepts>
- All Rosetta prep steps MUST be FULLY completed, load-project-context skill loaded and fully executed
</core_concepts>
<verification_process>
Run every checkpoint. Each must pass or have documented justification.
FILE EXISTENCE (non-empty, correct scope):
- TECHSTACK.md — detected technologies, frameworks, build tools
- CODEMAP.md — markdown headers, 3-4 levels, recursive children counts
- DEPENDENCIES.md — direct dependencies only (project, package, version)
- CONTEXT.md — business context only, no technical details
- ARCHITECTURE.md — technical architecture, references CODEMAP.md, no business context
- IMPLEMENTATION.md — current state, DRY references
- ASSUMPTIONS.md — unknowns with forward references
- AGENT MEMORY.md — self-defined purpose and initial entries
- Each document includes self-definition (purpose, content type, style)
INIT INTEGRITY:
- Init mode: exactly one of install, upgrade, plugin
- Composite workspace: top-level docs as registries if composite
- File inventory built before creation/update decisions
- Shell files: frontmatter + single ACQUIRE, zero inline logic
- load-project-context shell and bootstrap rule installed
- Shells match schema — no structural deviations, no absolute paths
- Only if phase 5 ran: docs/PATTERNS/ with INDEX.md; each pattern in 2+ locations; INDEX.md is consistent
CROSS-FILE CONSISTENCY:
- TECHSTACK frameworks appear in ARCHITECTURE
- CONTEXT, ARCHITECTURE, IMPLEMENTATION complement — no duplication
- skill
codingloaded and used as file creation reference - greppable headers used in all files
CONDITIONAL (if rules requested, N/A otherwise):
- KB SEARCHED for IDE/Agent rules — agent's built-in knowledge is obsolete, verify KB was queried
- Existing rules checked before creating new
- Root agents file uses mcp-files-mode.md template
- Tech-specific agent files created
- Local instructions with MoSCoW emphasis
- Weekly check mechanism with release version
- Subagents/commands initialized via KB instructions if supported
QUESTIONS:
- HIGH priority gaps addressed via targeted questions
CATCH-UP: For failed checkpoints — identify owning skill, execute, re-verify.
ASSUMPTIONS REVALIDATION:
- Resolved entries: mark with evidence
- Duplicates: keep most detailed
- Forward references: verify target files exist
- New assumptions: track any discovered during verification
</verification_process>
</execute_verification>
<next_steps step="9.3">
- If verification found failed checkpoints: list specific remediation actions
- Suggest next steps based on workspace state:
- Run coding workflow for first feature
- Review and customize generated docs
- Add project-specific patterns </next_steps>
<enforce_new_chat step="9.4">
- EMPHASIZE: MUST start a new chat session after init completes
- Current session context is polluted with init-specific state
- Mark state as COMPLETE in
agents/init-workspace-flow-state.md</enforce_new_chat>
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago Changed a59edde43b1a
- 1mo ago First seen · 128 lines · 16 tokens per session scan A d53ed68a775d
init-workspace-flow-verification is a command published in the GitHub repository griddynamics/rosetta (360 stars, last pushed today), licensed Apache-2.0. It adds 16 tokens to every session and 1,062 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other commands, from other repositories
qring-analyze
Run q-ring usage / optimization analysis — heavy keys, unused entries, decay risks.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.