Skill Funds

Funds Skill

Thin wrapper that delegates funding analysis work to funds-agent subagent.

IMPORTANT: This skill implements the skill-internal postflight pattern. After the subagent returns, this skill handles all postflight operations (status update, artifact linking, git commit) before returning. This eliminates the "continue" prompt issue between skill return and orchestrator.

Context References

Reference (do not load eagerly):

• Path: .claude/context/formats/return-metadata-file.md - Metadata file schema
• Path: .claude/context/patterns/postflight-control.md - Marker file protocol
• Path: .claude/context/patterns/file-metadata-exchange.md - File I/O helpers
• Path: .claude/context/patterns/jq-escaping-workarounds.md - jq escaping patterns (Issue #1132)

Note: This skill is a thin wrapper with internal postflight. Context is loaded by the delegated agent.

Trigger Conditions

This skill activates when:

• Task type is "present" with task_type="funds"
• /funds command with task number
• Present extension is available

Workflow Type Routing

This skill routes to funds-agent with one of four analysis modes:

Note: All modes follow the same status transition since they produce research-type output.

Input Parameters

Required Parameters

• task_number - Task number (must exist in state.json with language="present" and task_type="funds")
• session_id - Session ID from orchestrator

Optional Parameters

• topic - Topic for legacy standalone mode (--quick)
• mode - Analysis mode override (LANDSCAPE, PORTFOLIO, JUSTIFY, GAP)

Execution Flow

Stage 1: Input Validation

Validate required inputs:

• task_number - Must be provided and exist in state.json
• Task must have language="present" and task_type="funds"

# Lookup task
task_data=$(jq -r --argjson num "$task_number" \
  '.active_projects[] | select(.project_number == $num)' \
  specs/state.json)

# Validate exists
if [ -z "$task_data" ]; then
  return error "Task $task_number not found"
fi

# Extract fields
task_type=$(echo "$task_data" | jq -r '.task_type // "present"')
status=$(echo "$task_data" | jq -r '.status')
project_name=$(echo "$task_data" | jq -r '.project_name')
description=$(echo "$task_data" | jq -r '.description // ""')
task_type=$(echo "$task_data" | jq -r '.task_type // ""')

# Validate language is "present"
if [ "$task_type" = "present" | not ]; then
  return error "Task $task_number has language '$task_type', expected 'present'"
fi

# Validate task_type is "funds"
if [ "$task_type" = "funds" | not ]; then
  return error "Task $task_number has task_type '$task_type', expected 'funds'"
fi

# Validate status (only block terminal states)
if [ "$status" = "completed" ] || [ "$status" = "abandoned" ] || [ "$status" = "expanded" ]; then
  return error "Task is in terminal state [$status]"
fi

Stage 2 + Stage 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")
skill_name="skill-funds"
operation="research"

operation="research" (not "funds_analysis") is required here: update-task-status.sh's target_status vocabulary has no funds_analysis value, so this skill maps onto the plain research operation. The marker's operation field now reads "research" rather than "funds_analysis".

Stage 3a: Calculate Artifact Number

# Read next_artifact_number from state.json
next_num=$(jq -r --argjson num "$task_number" \
  '.active_projects[] | select(.project_number == $num) | .next_artifact_number // 1' \
  specs/state.json)

# Use current number for research artifacts (advances sequence)
artifact_number=$next_num
artifact_padded=$(printf "%02d" "$artifact_number")

# Increment next_artifact_number
bash .claude/scripts/state-write.sh \
  '(.active_projects[] | select(.project_number == $num)).next_artifact_number = $next' \
  --session-id "$session_id" \
  --argjson num "$task_number" \
  --argjson next "$((next_num + 1))"

Stage 4: Prepare Delegation Context

Extract forcing_data from task metadata and prepare delegation context:

# Extract forcing_data from task
forcing_data=$(echo "$task_data" | jq '.forcing_data // {}')
mode=$(echo "$forcing_data" | jq -r '.mode // "LANDSCAPE"')

Prepare delegation context for the subagent:

{
  "session_id": "sess_{timestamp}_{random}",
  "delegation_depth": 1,
  "delegation_path": ["orchestrator", "funds", "skill-funds"],
  "timeout": 3600,
  "task_context": {
    "task_number": N,
    "task_name": "{project_name}",
    "description": "{description}",
    "task_type": "present",
    "task_type": "funds"
  },
  "mode": "{selected_mode from forcing_data}",
  "forcing_data": "{forcing_data object}",
  "artifact_number": "{artifact_padded}",
  "metadata_file_path": "specs/{NNN}_{SLUG}/.return-meta.json"
}

Stage 4b: Read and Inject Format Specification

Read the summary format file and prepare it for injection into the subagent prompt:

format_content=$(cat .claude/context/formats/summary-format.md)

The format content will be included as a delimited section in the Stage 5 prompt.

Stage 5: Invoke Subagent

CRITICAL: You MUST use the Agent tool to spawn the subagent.

Required Tool Invocation:

Tool: Agent (NOT Skill, NOT Plan)
Parameters:
  - subagent_type: "funds-agent"
  - prompt: [Include task_context, delegation_context, mode, forcing_data, artifact_number, metadata_file_path,
             AND the format specification from Stage 4b]
  - description: "Execute funding analysis for task {N}"

Format Injection: Include the format specification from Stage 4b in the prompt as a clearly-delimited section:

<artifact-format-specification>
## CRITICAL: Summary Format Requirements

You MUST follow this format specification exactly when writing the implementation summary.
Non-compliance will be caught by postflight validation.

{format_content from Stage 4b}
</artifact-format-specification>

DO NOT use Skill(funds-agent) - this will FAIL.

The subagent will:

• Load funding analysis context files
• Parse forcing_data and mode
• Execute mode-specific analysis (web research, funder database queries)
• Create research report and optional XLSX/JSON outputs
• Write metadata to specs/{NNN}_{SLUG}/.return-meta.json
• Return a brief text summary (NOT JSON)

Stage 5a: Validate Subagent Return Format

If the subagent's text return parses as valid JSON, log a warning (v1 pattern instead of v2 file-based pattern). Non-blocking -- continue to read metadata file regardless.

Stage 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.

Stage 6: Read Metadata File

Read the metadata file:

metadata_file="specs/${padded_num}_${project_name}/.return-meta.json"

if [ -f "$metadata_file" ] && jq empty "$metadata_file" 2>/dev/null; then
    meta_status=$(jq -r '.status' "$metadata_file")
    artifact_path=$(jq -r '.artifacts[0].path // ""' "$metadata_file")
    artifact_type=$(jq -r '.artifacts[0].type // ""' "$metadata_file")
    artifact_summary=$(jq -r '.artifacts[0].summary // ""' "$metadata_file")
else
    echo "Error: Invalid or missing metadata file"
    meta_status="failed"
fi

Handle in_progress status: If metadata file shows status: "in_progress", the subagent was interrupted:

if [ "$meta_status" = "in_progress" ]; then
    partial_stage=$(jq -r '.partial_progress.stage // "unknown"' "$metadata_file")
    partial_details=$(jq -r '.partial_progress.details // ""' "$metadata_file")
    echo "Subagent interrupted at stage: $partial_stage"
    echo "Details: $partial_details"
fi

Stage 6a: Validate Artifact Content

If subagent status indicates success and artifact_path is non-empty, validate the artifact:

if [ "$meta_status" = "researched" ]; then
    if [ -n "$artifact_path" ] && [ -f "$artifact_path" ]; then
        echo "Validating artifact..."
        if ! bash .claude/scripts/validate-artifact.sh "$artifact_path" report --fix; then
            echo "WARNING: Artifact has format issues (non-blocking). Review output above."
        fi
    fi
fi

Stage 7: Update Task Status (Postflight)

Postflight Status Mapping:

Update state.json (if status changed to success):

if [ "$meta_status" = "researched" ]; then
  bash .claude/scripts/state-write.sh \
    '(.active_projects[] | select(.project_number == $num)) |= . + {
      status: $status,
      last_updated: $ts
    }' \
    --session-id "$session_id" \
    --argjson num "$task_number" \
    --arg ts "$(date -u +%Y-%m-%dT%H:%M:%SZ)" \
    --arg status "researched"
fi

Update TODO.md: Use Edit tool to change status marker to [RESEARCHED].

On partial/failed: Keep status at preflight level for resume.

Stage 8: Link Artifacts

Add artifact to state.json with summary.

IMPORTANT: Use two-step jq pattern to avoid Issue #1132 escaping bug.

if [ -n "$artifact_path" ]; then
    # Step 1: Filter out existing report artifacts (use "| not" pattern - Issue #1132)
    bash .claude/scripts/state-write.sh \
      '(.active_projects[] | select(.project_number == $num)).artifacts =
        [(.active_projects[] | select(.project_number == $num)).artifacts // [] | .[] | select(.type == "report" | not)]' \
      --session-id "$session_id" \
      --argjson num "$task_number"

    # Step 2: Add new artifact
    bash .claude/scripts/state-write.sh \
      '(.active_projects[] | select(.project_number == $num)).artifacts += [{"path": $path, "type": $type, "summary": $summary}]' \
      --session-id "$session_id" \
      --argjson num "$task_number" \
      --arg path "$artifact_path" \
      --arg type "$artifact_type" \
      --arg summary "$artifact_summary"
fi

Update TODO.md: Link artifact using count-aware format. Apply the four-case Edit logic from @.claude/context/patterns/artifact-linking-todo.md with field_name=**Research**, next_field=**Plan**.

Stage 9: Git Commit

Apply the research scope from .claude/context/standards/git-staging-scope.md — targeted staging, never a repo-wide add — then commit with session ID:

git add \
  "specs/${padded_num}_${project_name}/reports/" \
  "specs/${padded_num}_${project_name}/.return-meta.json" \
  "specs/TODO.md" \
  "specs/state.json"
git commit -m "task ${task_number}: complete funding analysis research

Session: ${session_id}

On commit failure: Non-blocking. Log the failure but continue with success response.

Stage 10: Cleanup

Follow @.claude/context/patterns/skill-postflight-flow.md's Stage 9 (cleanup):

skill_cleanup "$padded_num" "$project_name"

Stage 11: Return Brief Summary

Return a brief text summary (NOT JSON) based on analysis outcome.

Research Success:

Funding analysis research completed for task {N}:
- Mode: {mode} analysis
- Identified {count} funding opportunities
- Created report at specs/{NNN}_{SLUG}/reports/{MM}_funding-analysis.md
- Status updated to [RESEARCHED]
- Changes committed with session {session_id}

Next steps:
- Review the funding analysis report
- Run /plan {N} to create implementation plan
- Run /implement {N} to generate deliverables

Partial Return:

Funding analysis partially completed for task {N}:
- {completed_actions}
- {failed_action} failed: {reason}
- Partial report saved at specs/{NNN}_{SLUG}/reports/{MM}_funding-analysis.md
- Status remains [RESEARCHING] - run /funds {N} to continue

Error Handling

Input Validation Errors

Task not found:

Funds skill error for task {N}:
- Task not found in state.json
- Verify task exists with /task --sync
- No status changes made

Wrong language or task_type:

Funds skill error for task {N}:
- Task has language '{language}' / task_type '{task_type}', expected 'present' / 'funds'
- Use /funds "description" to create a new funding analysis task
- No status changes made

Metadata File Missing

If subagent didn't write metadata file:

1. Keep status at preflight level (researching)
2. Do not cleanup postflight marker
3. Report error to user with resume guidance

Funds skill error for task {N}:
- Subagent did not write metadata file
- Task remains [RESEARCHING] for resume
- Postflight marker preserved
- Run /funds {N} to retry

Git Commit Failure

Non-blocking error:

Funding analysis completed for task {N}:
- {analysis_results}
- [Warning] Git commit failed: {error}
- Manual commit recommended: review `git status --short`, then stage only the task directory
  (`specs/{padded}_{slug}/`) plus `specs/TODO.md` and `specs/state.json` before committing

Subagent Timeout

Return partial status (default 3600s timeout):

Funding analysis timed out for task {N}:
- Subagent exceeded timeout limit
- Partial progress: {partial_details}
- Status remains [RESEARCHING]
- Run /funds {N} to continue

MUST NOT (Postflight Boundary)

After the agent returns -- whether with status researched, partial, or failed -- this skill MUST proceed immediately to postflight (Stage 6). The skill MUST NOT:

1. Edit source/report files - All research work is done by agent
2. Run domain analysis or calculations - Analysis is agent work
3. Use MCP or WebSearch tools - Research tools are for agent use only
4. Analyze or grep source - Analysis is agent work
5. Write reports - Artifact creation is done by agent

PROHIBITION: If the subagent returned partial or failed status, the lead skill MUST NOT attempt to continue, complete, or "fill in" the subagent's work. Report the partial/failed status and let the user re-run /research to resume.

The postflight phase is LIMITED TO:

• Reading agent metadata file
• Calling update-task-status.sh for status updates (state.json + TODO.md)
• Linking artifacts in state.json
• Git commit
• Cleanup of temp/marker files

Reference: @.claude/context/standards/postflight-tool-restrictions.md

Return Format

This skill returns a brief text summary (NOT JSON). The JSON metadata is written to the file and processed internally.