CommunityResearch & Data Analysisgithub.com

samykabu/progress-report

Keep a live, deterministic progress report for a Markdown plan checklist or a Spec Kit feature. Collects task state from the plan file, hook events from Claude Code or Codex, and lane, blocker and decision facts you record, then renders a browser view and a Markdown export with no agent spent on upkeep.

What is progress-report?

progress-report is a Claude Code agent skill that keep a live, deterministic progress report for a Markdown plan checklist or a Spec Kit feature. Collects task state from the plan file, hook events from Claude Code or Codex, and lane, blocker and decision facts you record, then renders a browser view and a Markdown export with no agent spent on upkeep.

Works with✓Claude Code✓Codex CLI~Cursor
npx skills add https://github.com/samykabu/sanduq-skills/tree/HEAD/skills/agent-tools/skills/progress-report

Installed? Explore more Research & Data Analysis skills: affaan-m/uspto-database, affaan-m/scholar-evaluation, affaan-m/literature-review · View all 6 →

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentation

Progress report

Reporting is a deterministic projection. Never spawn, retain, message or poll an agent solely to maintain it. Record only missing assignment, decision or blocker facts.

Run the CLI from the repository root: progress below stands for python <skill dir>/scripts/progress.py (.claude/skills/progress-report, .agents/skills/progress-report, or $CLAUDE_PLUGIN_ROOT/skills/progress-report). Type the full command each time. Verbs print one line (refresh prints nothing); relay it.

When to use it

  • A plan file (Markdown with - [ ] / - [x] items) drives the work: plan mode, no Spec Kit needed.
  • A Spec Kit feature has specs/<feature>/tasks.md: feature mode; Workflow files are collected when present.

Plan mode quick start

  1. progress hooks install --host claude (or codex; add --scope user for a plugin install) once, for model, effort, context and child agents. Skip it and tasks are still tracked.
  2. progress init --plan docs/plan.md creates .progress/plan/ (see --name, --dir) and prints bind=<nonce>; the next hook binds (hooks install prints one if a report exists).
  3. Check items off in the plan file; the report follows. Progress never writes the plan.
  4. progress status gives the one-line state; the view is .progress/plan/.view/index.html.

Items keep a written id (T001); others get stable P001... ids. With one plan or feature active no target flag is needed; otherwise pass --plan NAME or --feature DIR.

Verbs

NeedCommand
Create a reportprogress init --plan PATH [--name N] [--dir D] or progress init (Spec Kit)
One-line stateprogress status
Assign a laneprogress lane L1 --tasks P001-P004 --owner ID --harness H --model M --effort E
Close a laneprogress lane L1 --close
Blockerprogress block P007 --by L2 (or P003, ext:"text"); progress unblock B4
Decisionprogress decide ask "question" --options a,b; progress decide made D2 b --by user
Noteprogress note "text" --target P005
Exportprogress export --format md
Hooksprogress hooks install|uninstall|status --host claude|codex [--scope project|user]; Codex also needs the project trusted and hooks approved

Rules

  1. Record a lane when you assign work (L#; host children C# appear on their own). Starts, completions and retries need no command.
  2. Never put secrets, absolute paths or transcript text in a note (200 characters at most).
  3. Unknown values stay unknown. Do not estimate model, effort, context or time.
  4. Errors read progress: error CODE; fix the cause, do not retry in a loop.

Details: references/verbs.md; contract docs/contracts.md; flags docs/commands.md.

Individual skills in this repo

This repo contains 4 individual skills — each has its own dedicated page.

samykabu/delegate-task

Delegate a task to an external agent CLI harness — Claude Code, OpenAI Codex, OpenCode, GitHub Copilot, or Pi — and get back a normalised result with status (successful / failed / abandoned), a summary, what actually changed on disk measured against git, and total input and output token usage.

samykabu/github-wiki-publish

Publish or update pages, reports, diagrams and assets in a GitHub repository Wiki, preserving existing content and verifying the published result. Use for GitHub Wiki publishing or sync requests, rather than GitHub Pages deployment or general documentation generation.

samykabu/illustrate

Draws diagrams, charts, and technical illustrations (44 types) as self-contained HTML with inline SVG, plus optional SVG/PNG/PDF export, in the project's tracked color and font theme. Use this skill whenever the user asks to draw, diagram, visualize, chart, map, sketch, or illustrate something, even without saying 'illustrate': architecture or system design, infrastructure, cloud or deployment topology, data flow, flowcharts, sequence, state machines, ER or database schema, UML class, dependency graphs, org charts, timelines, Gantt, swimlanes, process or workflow maps, user journeys, kanban, story maps, quadrants, Venn, pyramids, fishbone root-cause, Sankey, Wardley maps, bar, line, scatter, heatmap, waterfall, treemap, radar, or polar charts, and exploded or axonometric views. Also use to redraw .drawio, Mermaid (.mmd), or .excalidraw files, add a diagram to docs, a PR, or a spec, set up the illustration theme, or export a diagram. Not for photos or AI-generated raster images.

samykabu/user-manual

Create, update, audit and publish a standalone application User Manual with End User, Administrator and Technical editions, in English and optional Arabic. Use without Spec Kit to start a manual, document a feature or module, audit coverage, write API documentation (OpenAPI, AsyncAPI, GraphQL, webhooks), release notes and migration guides, plan and capture UI screenshots, set up private previews and versioned release…

Related Skills