B. User Prompting
When the workflow needs user input, prompt the user conversationally:
- Present options as a numbered list in your response text
- Ask the user to reply with their choice
- For multi-select, ask for comma-separated numbers
C. Tool Usage
Use these Augment tools when executing GSD workflows:
launch-process for running commands (terminal operations)str-replace-editor for editing existing filesview for reading files and listing directoriessave-file for creating new filesgrep for searching code (or use MCP servers for advanced search)web-search, web-fetch for web queriesadd_tasks, view_tasklist, update_tasks for task management
D. Subagent Spawning
When the workflow needs to spawn a subagent:
- Use the built-in subagent spawning capability
- Define agent prompts in
.augment/agents/ directory
Purpose: Diagnose failed or stuck workflows so the user can understand root cause and take corrective action.
Output: Forensic report saved to .planning/forensics/, presented inline, with optional issue creation.
User input:
- Problem description: {{GSD_ARGS}} (optional — will ask if not provided)
1---2name: gsd-forensics-33description: Post-mortem investigation for failed GSD workflows — diagnoses what went wrong.4---56<augment_skill_adapter>7## A. Skill Invocation8- This skill is invoked when the user mentions `gsd-forensics` or describes a task matching this skill.9- Treat all user text after the skill mention as `{{GSD_ARGS}}`.10- If no arguments are present, treat `{{GSD_ARGS}}` as empty.1112## B. User Prompting13When the workflow needs user input, prompt the user conversationally:14- Present options as a numbered list in your response text15- Ask the user to reply with their choice16- For multi-select, ask for comma-separated numbers1718## C. Tool Usage19Use these Augment tools when executing GSD workflows:20- `launch-process` for running commands (terminal operations)21- `str-replace-editor` for editing existing files22- `view` for reading files and listing directories23- `save-file` for creating new files24- `grep` for searching code (or use MCP servers for advanced search)25- `web-search`, `web-fetch` for web queries26- `add_tasks`, `view_tasklist`, `update_tasks` for task management2728## D. Subagent Spawning29When the workflow needs to spawn a subagent:30- Use the built-in subagent spawning capability31- Define agent prompts in `.augment/agents/` directory32</augment_skill_adapter>3334<objective>35Investigate what went wrong during a GSD workflow execution. Analyzes git history, `.planning/` artifacts, and file system state to detect anomalies and generate a structured diagnostic report.3637Purpose: Diagnose failed or stuck workflows so the user can understand root cause and take corrective action.38Output: Forensic report saved to `.planning/forensics/`, presented inline, with optional issue creation.39</objective>4041<execution_context>42@/home/delorenj/code/bhappy/.augment/get-shit-done/workflows/forensics.md43</execution_context>4445<context>46**Data sources:**47- `git log` (recent commits, patterns, time gaps)48- `git status` / `git diff` (uncommitted work, conflicts)49- `.planning/STATE.md` (current position, session history)50- `.planning/ROADMAP.md` (phase scope and progress)51- `.planning/phases/*/` (PLAN.md, SUMMARY.md, VERIFICATION.md, CONTEXT.md)52- `.planning/reports/SESSION_REPORT.md` (last session outcomes)5354**User input:**55- Problem description: {{GSD_ARGS}} (optional — will ask if not provided)56</context>5758<process>59Read and execute the forensics workflow from @/home/delorenj/code/bhappy/.augment/get-shit-done/workflows/forensics.md end-to-end.60</process>6162<success_criteria>63- Evidence gathered from all available data sources64- At least 4 anomaly types checked (stuck loop, missing artifacts, abandoned work, crash/interruption)65- Structured forensic report written to `.planning/forensics/report-{timestamp}.md`66- Report presented inline with findings, anomalies, and recommendations67- Interactive investigation offered for deeper analysis68- GitHub issue creation offered if actionable findings exist69</success_criteria>7071<critical_rules>72- **Read-only investigation:** Do not modify project source files during forensics. Only write the forensic report and update STATE.md session tracking.73- **Redact sensitive data:** Strip absolute paths, API keys, tokens from reports and issues.74- **Ground findings in evidence:** Every anomaly must cite specific commits, files, or state data.75- **No speculation without evidence:** If data is insufficient, say so — do not fabricate root causes.76</critical_rules>
delorenj