Deck Implement Skill
Routes deck-specific implementation requests to the deck-builder-agent, generating Slidev pitch decks from plans created by skill-deck-plan and research from skill-deck-research.
Context Pointers
Reference (do not load eagerly):
- Path:
.claude/context/formats/subagent-return.md - Purpose: Return validation
- Load at: Subagent execution only
Note: This skill is a thin wrapper. Context is loaded by the delegated agent, not this skill.
Trigger Conditions
This skill activates when:
Direct Invocation
- An implement-phase dispatch (via
/orchestrate) on a task withlanguage: founderandtask_type: deck - Extension routing lookup finds
routing.implement["founder:deck"]
Task-Type-Based Routing
- Task type is "founder" AND task_type is "deck"
- An implement-phase dispatch for {N} where task {N} has language="founder", task_type="deck"
When NOT to trigger
Do not invoke for:
- Tasks with other task_types (market, analyze, strategy, legal, project, sheet, finance)
- Tasks with other language types (general, meta, neovim, etc.)
- Quick mode operations (
--quickflag) - Tasks in [NOT STARTED] status (need plan first)
- Tasks already [COMPLETED]
Execution
1. Input Validation
Validate inputs from delegation context:
task_number- Required, integerplan_path- Required, path to implementation planresume_phase- Optional, phase number to resume fromsession_id- Required, string
# Validate task_number is present
if [ -z "$task_number" ]; then
return error "task_number is required"
fi
# Validate session_id is present
if [ -z "$session_id" ]; then
return error "session_id is required"
fi
# Validate plan_path exists
if [ -z "$plan_path" ] || [ ! -f "$plan_path" ]; then
return error "plan_path is required and must exist. Run /orchestrate first."
fi
2 + 3. Preflight Status Update and Postflight Marker
Source skill-base.sh once, then follow @.claude/context/patterns/skill-preflight-flow.md in
full for Stage 2 (preflight status update) and Stage 3 (marker creation):
source .claude/scripts/skill-base.sh
padded_num=$(printf "%03d" "$task_number")
project_name=$(jq -r --argjson num "$task_number" \
'.active_projects[] | select(.project_number == $num) | .project_name' \
specs/state.json)
task_dir="specs/${padded_num}_${project_name}"
mkdir -p "$task_dir"
skill_name="skill-deck-implement"
operation="implement"
4. Context Preparation
Extract template palette from the plan file:
# Extract template palette from plan (look for "palette:" or "template:" in plan metadata)
template_palette=$(grep -oP '(?:palette|template):\s*\K[\w-]+' "$plan_path" | head -1)
template_palette="${template_palette:-dark-blue}"
# Extract forcing_data from state.json if available
forcing_data=$(jq -r --argjson num "$task_number" \
'.active_projects[] | select(.project_number == $num) | .forcing_data // null' \
specs/state.json)
Prepare delegation context for agent:
{
"task_context": {
"task_number": 234,
"project_name": "seed_round_pitch_deck",
"description": "Seed round pitch deck",
"task_type": "founder",
"task_type": "deck"
},
"plan_path": "specs/234_seed_round_pitch_deck/plans/01_deck-plan.md",
"resume_phase": 1,
"output_dir": "strategy/",
"template_palette": "dark-blue",
"forcing_data": {
"purpose": "INVESTOR",
"source_materials": ["task:233"]
},
"metadata_file_path": "specs/234_seed_round_pitch_deck/.return-meta.json",
"metadata": {
"session_id": "sess_{timestamp}_{random}",
"delegation_depth": 2,
"delegation_path": ["orchestrator", "implement", "skill-deck-implement"]
}
}
5. Invoke Agent
CRITICAL: You MUST use the Agent tool to spawn the agent.
Required Tool Invocation:
Tool: Agent (NOT Skill, NOT Plan)
Parameters:
- subagent_type: "deck-builder-agent"
- prompt: [Include task_context, plan_path, resume_phase, output_dir, template_palette, forcing_data, metadata]
- description: "Generate Slidev pitch deck from plan and research"
The agent will:
- Load plan and research report
- Load theme config from
.context/deck/themes/ - Assemble
slides.mdfrom library content with slot filling - Compose CSS styles and copy Vue components
- Attempt non-blocking
slidev exportfor PDF - Write-back new content to library
- Create summary in task directory
- Write metadata file for postflight consumption
- Return brief text summary
Note: Slidev export (PDF generation) is optional. Task completes successfully
even if slidev is not installed or PDF generation fails. The slides.md source is preserved.
5b. Self-Execution Fallback
CRITICAL: If you performed the work above WITHOUT using the Agent tool (i.e., you read files,
wrote artifacts, or updated metadata directly instead of spawning a subagent), you MUST write a
.return-meta.json file now before proceeding to postflight. Use the schema from
return-metadata-file.md with the appropriate status value for this operation.
If you DID use the Agent tool, skip this stage -- the subagent already wrote the metadata.
Postflight (ALWAYS EXECUTE)
The following stages MUST execute after work is complete, whether the work was done by a subagent or inline (Stage 5b). Do NOT skip these stages for any reason.
6. Read Metadata File
Read the metadata file:
metadata_file="specs/${padded_num}_${project_name}/.return-meta.json"
metadata=$(cat "$metadata_file")
status=$(echo "$metadata" | jq -r '.status')
7. Postflight Status Update
If agent succeeded (status == "implemented"):
# Update state.json to completed
bash .claude/scripts/state-write.sh \
'(.active_projects[] | select(.project_number == $num)) += {
status: "completed",
last_updated: $ts
}' \
--session-id "$session_id" \
--argjson num "$task_number" \
--arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)"
# Link artifacts in state.json (deck files and summary)
artifacts=$(echo "$metadata" | jq '.artifacts')
for i in $(seq 0 $(($(echo "$artifacts" | jq 'length') - 1))); do
artifact_type=$(echo "$artifacts" | jq -r ".[$i].type")
artifact_path=$(echo "$artifacts" | jq -r ".[$i].path")
artifact_summary=$(echo "$artifacts" | jq -r ".[$i].summary")
bash .claude/scripts/state-write.sh \
'(.active_projects[] | select(.project_number == $num)).artifacts += [{
type: $type,
path: $path,
summary: $summary
}]' \
--session-id "$session_id" \
--argjson num "$task_number" \
--arg type "$artifact_type" \
--arg path "$artifact_path" \
--arg summary "$artifact_summary"
done
Update TODO.md status marker to [COMPLETED], add Completed date, and link summary artifact per @.claude/context/patterns/artifact-linking-todo.md with field_name=**Summary**, next_field=**Description**.
If partial (status == "partial"):
- Keep status as "implementing"
- Note resume point in metadata
8. Git Commit
Apply the implement scope from .claude/context/standards/git-staging-scope.md — targeted
staging, never a repo-wide add:
stage_paths=("${task_dir}/" "specs/TODO.md" "specs/state.json")
metadata_file="${task_dir}/.return-meta.json"
while IFS= read -r f; do
[ -n "$f" ] && stage_paths+=("$f")
done < <(jq -r '.modified_files[]? // empty' "$metadata_file" 2>/dev/null)
bash .claude/scripts/git-commit-scoped.sh \
--message "task {N}: complete implementation" \
--session "{session_id}" \
--honest-index-rows {N} \
-- "${stage_paths[@]}"
9. Cleanup and Return
Follow @.claude/context/patterns/skill-postflight-flow.md's Stage 9 (cleanup):
skill_cleanup "$padded_num" "$project_name"
Return brief text summary to caller.
MUST NOT (Postflight Boundary)
After the agent returns -- whether with status implemented, partial, or failed -- this skill MUST proceed immediately to postflight (Stage 6). The skill MUST NOT:
- Edit slides/deck files - All deck generation is done by agent
- Run slidev commands - Export/build is done by agent
- Use MCP tools - Domain tools are for agent use only
- Analyze or grep source - Analysis is agent work
- Write summary/reports - Artifact creation is done by agent
See context/standards/postflight-tool-restrictions.md § "MUST NOT Section Template"
for the full PROHIBITION clause (the lead skill must not continue, complete, or "fill in" a
partial/failed subagent's work; report the status and let the user re-run the implement-phase dispatch) and the
postflight-phase allowed-operations list (read metadata, state.json/TODO.md updates, artifact
linking, git commit, cleanup).
Return Format
Brief text summary (NOT JSON).
Expected successful return (with PDF):
Deck implementation completed for task {N}:
- Theme: {theme_name}, Pattern: {pattern_name}
- Slides populated: {M}/{total} from research
- Slidev source: strategy/{slug}-deck/slides.md
- PDF: strategy/{slug}-deck/{slug}-deck.pdf
- Summary: specs/{NNN}_{SLUG}/summaries/01_{short-slug}-summary.md
- Status updated to [COMPLETED]
- Changes committed with session {session_id}
- Next: Review deck and fill remaining [TODO:] markers
Expected successful return (without PDF):
Deck implementation completed for task {N}:
- Theme: {theme_name}, Pattern: {pattern_name}
- Slides populated: {M}/{total} from research
- Slidev source: strategy/{slug}-deck/slides.md
- PDF: skipped (slidev not installed)
- Summary: specs/{NNN}_{SLUG}/summaries/01_{short-slug}-summary.md
- Status updated to [COMPLETED]
- Changes committed with session {session_id}
- Next: Install slidev for PDF export, or review slides.md source
Expected partial return:
Deck implementation partially completed for task {N}:
- Template selected but content generation incomplete
- Resume phase: {resume_phase}
- Status remains [IMPLEMENTING]
- Next: Run /orchestrate {N} to resume
Error Handling
Session ID Missing
Return immediately with failed status.
Plan Not Found
Return error with guidance to run the plan phase first.
Task Not Found
Return error with guidance to check task number.
Agent Errors
Pass through the agent's error return verbatim.
Slidev/PDF Errors
PDF export failures do NOT block task completion:
- Slidev not installed: Task completes with
slides.mdsource only - Export error: Keep
slides.mdfor debugging, task completes - PDF empty: Keep
slides.md, task completes
Postflight should check metadata.pdf_generated to determine what artifacts to report.