* Add doublecheck plugin: three-layer verification pipeline for AI output
Adds a new plugin that helps users verify AI-generated content before
acting on it. Designed for sensitive contexts (legal, medical, financial,
compliance) where hallucinations carry real consequences.
Three verification layers:
- Self-Audit: extracts verifiable claims, checks internal consistency
- Source Verification: web searches per claim, produces URLs for human review
- Adversarial Review: assumes errors exist, checks hallucination patterns
Supports persistent mode (auto-verifies every factual response inline)
and one-shot mode (full report on specific text). Confidence ratings:
VERIFIED, PLAUSIBLE, UNVERIFIED, DISPUTED, FABRICATION RISK.
Includes:
- Skill (skills/doublecheck/) with bundled report template
- Agent (agents/doublecheck.agent.md) for interactive verification
- Plugin package (plugins/doublecheck/) bundling both
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address review: fix tools YAML format, remove materialized artifacts
- Fix tools frontmatter in agents/doublecheck.agent.md to use standard
YAML list format instead of flow sequence with trailing comma
- Remove plugins/doublecheck/agents/ and plugins/doublecheck/skills/
from tracking; these paths are in .gitignore as CI-materialized
artifacts that should not be committed
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Add napkin plugin: visual whiteboard collaboration for Copilot CLI
Napkin opens an interactive HTML whiteboard in the user's browser where
they can draw, sketch, and add sticky notes. When ready, they click
'Share with Copilot' which exports a PNG snapshot. The Copilot CLI agent
reads the PNG via the view tool, and the multimodal AI model interprets
the drawings, spatial layout, and text content — responding
conversationally as a collaborator.
Built for non-technical users: lawyers, PMs, business stakeholders,
designers, and anyone who thinks better visually.
Includes:
- Self-contained HTML whiteboard (zero external dependencies)
- Shape recognition (wobbly shapes snap to clean versions)
- Freehand drawing, sticky notes, arrows, text labels
- Auto-save to localStorage
- SKILL.md with agent instructions
- SVG visual guides for step-by-step README documentation
- Plugin manifest for Copilot CLI installation
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix review feedback: roundRect compat, cross-platform shortcuts, eraser undo
- Add safeRoundRect() fallback for browsers without CanvasRenderingContext2D.roundRect()
- Update undo/redo button titles to show Ctrl/Cmd instead of Mac-only Cmd
- Fix eraser undo by capturing state before first erase mutation per gesture
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Update skills/napkin/templates/napkin.html
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Update plugins/napkin/README.md
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Update skills/napkin/templates/napkin.html
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Update skills/napkin/templates/napkin.html
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Update skills/napkin/templates/napkin.html
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Add missing Line (L) and Eraser (E) keyboard shortcuts to README
The in-app shortcuts panel and keyMap handler both support L for Line
and E for Eraser, but the README keyboard shortcuts table was missing
both entries. Adds them to match the actual implementation.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Combine docs/ and templates/ into assets/ per agentskills.io spec
Moves skills/napkin/docs/*.svg and skills/napkin/templates/napkin.html
into skills/napkin/assets/ to align with the Agent Skills specification.
Updates all path references in SKILL.md, README.md, and generated docs.
Addresses review feedback from PR #929.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Dan Velton <dvelton@Dans-MacBook-Pro.local>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Add automate-this plugin: screen recording to automation scripts
New plugin that analyzes screen recordings of manual processes and
proposes working automation scripts at multiple complexity tiers.
Features:
- Frame extraction via ffmpeg for visual process reconstruction
- Optional audio transcription via Whisper for narrated recordings
- Environment fingerprinting to constrain proposals to installed tools
- Application-specific automation strategies (browser, spreadsheet,
email, terminal, file management, macOS-native)
- Three-tier proposal system (quick win, full script, scheduled automation)
- Dry-run support for safe testing of all proposals
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Update skills/automate-this/SKILL.md
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Address code review feedback on automate-this plugin
- Fix privacy section: clarify that frames/audio are processed locally
but sent to the Copilot model for analysis (not purely local)
- Replace 'which' with 'command -v' for reliable tool detection with
explicit NO_FFMPEG/NO_WHISPER sentinels
- Use per-run secure temp directory (mktemp -d, mode 0700) instead of
fixed /tmp/ paths to prevent other users reading extracted frames
- Add -y flag and -loglevel warning to ffmpeg calls to prevent
overwrite prompts and avoid piping hiding exit codes
- Add whisper-cpp CLI invocation path so users who install only
whisper-cpp get working transcription
- Fix grouped command syntax in environment fingerprint to prevent
false 'not installed' fallthrough
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
- Make browser tester generic to support for chrome devotols mcp, playwright, agentic browser tools.
- Add Team lead and energetci peronsality to Orchestrator
- Add progress updates between phases/ waves
* feat: Prd/ steer support
- Add supprot for PRD
- Vscode steer/ queue support
- Consistent artifacts
- Improved parallel running; for researchers too
* chore: improve prd update support
* chore: Make reviewer use prd for compaince
* chore: imrpvoe websearch in researcher
* fix(gem-team): revert gem-team plugin version from 1.5.0 to 1.2.0
Noob Mode is a plain-English translation layer for non-technical Copilot
CLI users (PMs, business stakeholders, designers, and anyone new to the
command line). When activated, it translates every approval prompt, error
message, and technical output into clear, jargon-free English with
color-coded risk indicators.
Includes:
- Standalone skill (skills/noob-mode/) with SKILL.md and bundled
glossary (100+ terms) and before/after examples
- Plugin (plugins/noob-mode/) for installation via
copilot plugin install noob-mode@awesome-copilot
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Modified tools to make the agent more efficient
* Added skill for importing Azure Infrastructure as Code
* Updated readme
* Some wording improvements
* Added / to skills path endings
* ran npm start
- Bump plugin version to 1.1.0 in marketplace and plugin.json
- Rename agent from "Chrome Tester" to "Browser Tester" in plugin.json
- Update agent description to focus on browser automation tools instead of Chrome DevTools
- Add symlink for the Browser Tester agent in the plugin's agents directory
