Generate Quarto and HTML reports of local Codex session cost, token usage by model, project, coding vs cowork cost, weekly trends, annual projections, and top expensive sessions.

2026-07-010

smoketest

chase-seibert/chase-sidekick

Check that core Sidekick integrations can authenticate and read common restricted files

2026-06-300

weekly-report

chase-seibert/chase-sidekick

Generate summary of 1:1 and meeting notes organized by audience, wins, and kudos

2026-06-300

capture-work-todos

chase-seibert/chase-sidekick

Capture missing TODOs from recent work context. Use when Codex needs to scan the user's Slack DMs with their manager or direct reports, manager/direct-report 1:1 docs, and leadership meeting notes for direct asks or action items assigned to the user, deduplicate against the user's configured todo app, and create missing tasks with source links and searchable metadata.

2026-06-250

interview-scorecard

chase-seibert/chase-sidekick

Draft concise interview scorecards from pasted long-form interview notes, nearby Google Calendar interview context, and memory/miclog/ transcript excerpts. Use when writing hiring feedback, interview scorecards, spikes/troughs, or candidate evaluation summaries.

2026-06-250

jira-hours

chase-seibert/chase-sidekick

Calculate a team's average Jira hours from Rovo Jira completed work, with optional Workday Approved Time Off absence adjustments for a date window and team roster.

2026-06-250

name	epic-assignment
description	Assign active Epics to Roadmap Initiatives based on recent activity
argument-hint	<root-issue> [project] [issue-type] [days]
auto-approve	true

Epic Assignment Agent

This agent identifies recently active Epics without parents and generates recommendations for linking them to appropriate parent issues.

Purpose

Ensures active Epics are properly linked to Parent Epics for better project tracking and visibility. Many Epics may be created without parents, making it difficult to track which roadmap item they support.

Prerequisites

This agent requires the following sections in @AGENTS.override.md:

Teams Section

## Teams

### My Teams
- Team Name, manager Name, JIRA Project KEY1
- Team Name, manager Name, JIRA Project KEY2

Projects Section

## Projects

The Roadmap issues are represented by the JIRA jql query:

project = "PROJECT" and issuetype = "Roadmap Issue Type"

Workflow

Step 1: Extract Configuration

You already have context form @AGENTS.override.md for

JIRA project for my teems
JQL for roadmap issues

Step 2: Query Recent Activity

Query JIRA for all issues updated in the specified time period (default 30 days):

Use Atlassian Rovo MCP JQL search first:

project IN ("KEY1", "KEY2", "KEY3") AND updated >= -30d

Fall back to python3 -m sidekick.clients.jira query ... only when Rovo is unavailable or lacks the needed fields.

Note: escape project keys with double quotes because they can be reserved words

Fields to extract: key, summary, issuetype, parent, labels, description, updated

Step 3: Extract Epics

From query results, identify Epics:

Epics: Issues where issuetype.name == "Epic"
Non-Epics: get parent Epics by looking at the parent field
Create a set of distinct Epics

From THOSE Epics, create two collections:

Epics without parents: parent field is None or empty
Epics with parents: parent field exists (for reference)

Group by project for the analysis report.

Step 4: Query Roadmap Initiatives and Build Hierarchies

Use Atlassian Rovo MCP first to query for all relevant potential parents, starting at a root issue. Use the /jira-roadmap skill only as a fallback when Rovo is unavailable or local recursive tree output is specifically useful.

This captures:

All Epics currently under this initiative
The complete structure of work already organized under this initiative

Step 5: Match Epics to Roadmap Initiatives

For each Epic without a parent, analyze potential matches based on:

Label overlap: Shared labels between Epic and Initiative (or its child Epics)
Keyword similarity: Keyword matches in summaries and descriptions (Epic vs Initiative + child Epics)
Theme alignment: Does this Epic fit the theme/focus area of existing child Epics?
Sibling similarity: Does this Epic resemble existing Epics under the Initiative?

Scoring system:

Label match: +15 points per shared label
Summary keyword overlap: +10 points per keyword
Description keyword overlap: +5 points per keyword
Sibling Epic similarity: +8 points per keyword match with existing child Epics

Assign confidence levels:

High: Score >= 30 (strong label/keyword overlap)
Medium: Score >= 15 (moderate overlap)
Low: Score >= 5 (weak overlap, manual review needed)
No Match: Score < 5

Step 6: Generate Analysis Report

Save to memory/epic_assignment_analysis.md:

# Epic Assignment Analysis
**Generated**: [timestamp]
**Activity Period**: Last [N] days

## Summary
- Total issues updated: X
- Epics identified: Y
- Epics without parents: Z
- Epics with parents: W

## Epics by Project

### Project: KEY1

#### Epics without parents
- [KEY1-123](https://company.atlassian.net/browse/KEY1-123): Epic Summary
  - Labels: label1, label2
  - Updated: YYYY-MM-DD

#### Epics with parents
- [KEY1-456](https://company.atlassian.net/browse/KEY1-456): Epic Summary
  - Parent: [INIT-789](https://company.atlassian.net/browse/INIT-789) - Initiative Name
  - Updated: YYYY-MM-DD

Step 7: Generate Recommendations Report

Save to memory/epic_assignment_recommendations.md:

# Epic Assignment Recommendations
**Generated**: [timestamp]

## Proposed Assignments

### High Confidence Matches
- **[KEY1-123](https://company.atlassian.net/browse/KEY1-123)**: Epic Summary
  - **Recommended Parent**: [INIT-456](https://company.atlassian.net/browse/INIT-456) - Initiative Summary
  - **Confidence**: High (score: 45)
  - **Reasoning**: Shared labels: label1, label2 • Keyword overlap with initiative • Similar to existing child Epics: KEY1-100, KEY1-105
  - **Existing Epics under this Initiative**: 3 Epics with similar themes

### Medium Confidence Matches
- **[KEY2-789](https://company.atlassian.net/browse/KEY2-789)**: Epic Summary
  - **Recommended Parent**: [INIT-234](https://company.atlassian.net/browse/INIT-234) - Initiative Summary
  - **Confidence**: Medium
  - **Reasoning**: [Explanation with caveats]

### Low Confidence / Needs Manual Review
- **[KEY3-456](https://company.atlassian.net/browse/KEY3-456)**: Epic Summary
  - **Possible Parents**:
    - [INIT-111](https://company.atlassian.net/browse/INIT-111) - Initiative A
    - [INIT-222](https://company.atlassian.net/browse/INIT-222) - Initiative B
  - **Confidence**: Low
  - **Reasoning**: [Multiple potential matches, manual review needed]

### No Match Found
- **[KEY4-789](https://company.atlassian.net/browse/KEY4-789)**: Epic Summary
  - **Reasoning**: No Roadmap Initiatives with relevant labels or keywords

## Summary Statistics
- High confidence: X
- Medium confidence: Y
- Low/manual review: Z
- No match: W

Step 8: User Confirmation

Output to console:

I've analyzed X Epics without parents and generated recommendations.

Summary:
- High confidence matches: X
- Medium confidence matches: Y
- Needs manual review: Z
- No match found: W

Reports saved to:
- memory/epic_assignment_analysis.md
- memory/epic_assignment_recommendations.md

Would you like me to:
1. Proceed with HIGH confidence assignments only? (type "high")
2. Proceed with HIGH + MEDIUM confidence assignments? (type "all")
3. Skip updates and just review? (type "skip")

Wait for user response before proceeding.

Step 9: Execute Updates

Based on user selection, update JIRA for each confirmed Epic assignment:

Use Atlassian Rovo MCP update issue first to set the parent field, after the user has confirmed the update set. Fall back to python3 -m sidekick.clients.jira update-issue KEY1-123 '{"parent": {"key": "INIT-456"}}' only when Rovo is unavailable or cannot update the needed field.

Error Handling:

Catch and log all errors without stopping
Continue processing remaining Epics if one fails
Report all failures in the changelog

Step 10: Generate Change Log

After executing updates, save to memory/epic_assignment_changelog.md:

# Epic Assignment Change Log
**Executed**: [timestamp]
**Execution Mode**: [High only / High + Medium / Skipped]

## Successfully Updated
- [KEY1-123](https://company.atlassian.net/browse/KEY1-123): Epic Summary
  - **Parent Set**: [INIT-456](https://company.atlassian.net/browse/INIT-456) - Initiative Summary
  - **Confidence**: High
  - **Time**: [timestamp]

## Failed Updates
- [KEY2-789](https://company.atlassian.net/browse/KEY2-789): Epic Summary
  - **Attempted Parent**: [INIT-234](https://company.atlassian.net/browse/INIT-234)
  - **Error**: [error message]
  - **Time**: [timestamp]

## Not Updated (Below Threshold)
- [KEY3-456](https://company.atlassian.net/browse/KEY3-456): Epic Summary
  - **Reason**: Medium confidence, user chose "high" only
  - **Recommended Parent**: [INIT-111](https://company.atlassian.net/browse/INIT-111)

## Summary
- Successfully assigned: X
- Failed: Y
- Skipped: Z

Output Files

All reports are saved to the memory/ directory:

epic_assignment_analysis.md - Initial analysis of all Epics (with and without parents)
epic_assignment_recommendations.md - Matching recommendations with confidence levels
epic_assignment_changelog.md - Log of updates made (created only after execution)

Each report must end with exactly:

This report generated using [chase-sidekick](https://github.com/chase-seibert/chase-sidekick) and the [epic-assignment skill](https://github.com/chase-seibert/chase-sidekick/tree/main/.agents/skills/epic-assignment).

Edge Cases

Epic already has parent: Appears in "Epics with parents" section only
Reserved JQL keyword: Project names that are keywords must be quoted
Update permission failure: Logged as error, processing continues
No Epics without parents: Report states "No Epics without parents found"
No matching Roadmap Initiatives: Epics appear in "No Match Found" section
Ambiguous matches: Multiple initiatives listed as options in Low Confidence section

Verification Steps

After running the agent:

Review analysis report: Check that Epics are correctly categorized by project
Review recommendations: Verify confidence levels match reasoning quality
Test dry-run: Select "skip" to generate reports without making changes
Review changelog: After updates, verify changelog accurately reflects changes
Check JIRA: Manually verify a sample of updated Epics show correct parent links

Notes

Hierarchy-based matching: Uses get_issue_hierarchy() to analyze existing Epic structure under each Initiative, enabling "sibling similarity" matching - if an Epic looks like existing Epics under an Initiative, it's likely a good fit
Agent uses temporary files for intermediate JSON if needed
Only final reports are saved to memory/ directory
Agent provides verbose console output showing progress
All console output uses relative paths for readability
JIRA parent field syntax: {"parent": {"key": "PARENT-KEY"}}
Default activity window is 30 days (customizable via argument)
Roadmap Initiatives filtered to 90-day activity window to avoid stale initiatives
Matching quality: Hierarchy analysis significantly improves matching accuracy by considering the context of existing work organization, not just keyword overlap