Automates unfinished STV trace scenarios from docs/*/trace.md through stv:work and quality gates.
npx skills add https://github.com/2lab-ai/oh-my-claude --skill do-work-2lab-ai
Please help me install this Agent Skill. Skill: do-work Source: https://github.com/2lab-ai/oh-my-claude/tree/main/plugins/stv/skills/do-work Command: npx skills add https://github.com/2lab-ai/oh-my-claude --skill do-work-2lab-ai
This Skill automates implementing unfinished STV trace scenarios by selecting ready items from trace.md, performing the required code changes, enforcing quality gates, and looping until scenarios are complete or human input is required. It removes manual orchestration of multi-file edits, verification, and commit cycles so features are implemented reliably and consistently.
Run the do-work skill to autonomously implement unfinished scenarios listed in docs/*/trace.md and progress them until the File Map is complete.
High-intent search queries and answers about installing and using this skill.
Automating trace scenario implementation involves scanning docs/*/trace.md, extracting and deduplicating File Map entries, bundling scenarios into execution-sized chunks, and running an autonomous stv:work loop that enforces quality gates and commits changes.
STV trace scenarios are documented in docs/*/trace.md files listing persisted files and side effects. You need this process when unfinished scenarios require iterative code implementation, gap detection, and spec re-verification across a repository.
Yes, your project requires docs/*/trace.md files where scenarios list persisted files and side effects. The automation extracts File Map entries from these docs to bundle work, enforce completion gates, run gap detection, and execute test, build, and lint checks.
Handling gap detection and spec re-verification requires an autonomous execution loop that invokes stv:work, applies File Map completion gates, runs fixes, and commits changes only when spec requirements are fully satisfied.
User intervention is required when context thresholds for autonomous decisions are exceeded. The process manages context budget, applies decision safety thresholds, escalates to the user, and produces structured session reports for handoff when scenarios cannot complete automatically.
2lab-aiImplement code scenarios from trace.md until contract tests pass.
pbsRun SCC, VTT, and DFXP compliance checks in sequence and generate dated reports.
canpok1Run build, lint:fix, and test commands to resolve code quality issues.
duranserkanOrchestrate autonomous development cycles with discovery, planning, execution, and verification.
