orca-background-mode
Run bounded unattended ORCA Framework progress loops with clear autonomy levels, loop guards, permission handling, and resumable receipts.
Skill by henryvn27
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/henryvn27-orca-background-mode
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/orca-background-mode
What's inside
- Skill name
orca-background-mode- Description
- Run bounded unattended ORCA Framework progress loops with clear autonomy levels, loop guards, permission handling, and resumable receipts.
- License
- MIT
- Instructions
- 61 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# ORCA Framework Background Mode ## What This Skill Is A guarded autonomy workflow for unattended progress. ## Trigger Use when the user explicitly asks ORCA Framework to keep making progress while they are away or asleep. ## Do Not Trigger Do not use for vague "keep improving things" requests without scope, limits, or stop conditions. ## Required Inputs - task, milestone, goal, or work item ## Optional Inputs - autonomy level - risk tier - time budget - step budget - allowed scope ## Exact Workflow 1. Define or read a background run contract. 2. Classify the task type and risk tier. 3. Set hard limits for time and steps. 4. Choose the lowest sufficient autonomy level. 5. Require each iteration to add new evidence, structure, or artifact progress. 6. Trigger circuit breakers on repeated failures, repeated blocked permissions, or semantic repetition. 7. Stop with one final state and emit a background run receipt. 8. Make the run resumable through status, receipts, and prior contract state. ## Expected Outputs - background run contract - background run plan - background run receipt - blocked-permission note when relevant ## Quality Bar The run should make real bounded progress or stop cleanly. It should never rely on "maybe the next retry will work" as its core strategy. ## Common Failure Modes - unattended scope creep - retrying the same blocked permission - rewriting the same plan without new evidence - taking Tier 3 actions in unattended mode - ending without a final state ## Relationship To Other ORCA Framework Skills And Commands Works with goal mode, runtime adaptation, receipts, lineage, status, checkpoints, and stage budgets.
From henryvn27/orca-framework, MIT.
More from henryvn27/orca-framework
- 0.613score
orca-build
henryvn27
Implement planned changes with scoped edits, verification, and respect for existing project patterns.
SkillAFreshness: A★ 5Claude Code +9 - 0.609score
orca-blind-qa
henryvn27
Run a zero-context first-look QA pass and produce a realistic first-impression report.
SkillAFreshness: A★ 5Claude Code +9 - 0.609score
orca-context-brief
henryvn27
Create bounded context packets for briefed QA and layered retesting.
SkillAFreshness: A★ 5Claude Code +9 - 0.528score
orca-ci
henryvn27
Triage failing CI checks, extract actionable failure context, and route to the right fix path.
SkillCFreshness: C★ 5Claude Code +9 - 0.528score
orca-corpus-support
henryvn27
Use optional persistent corpus support only when the user explicitly wants durable reusable vault grounding.
SkillCFreshness: C★ 5Claude Code +9 - 0.526score
orca-attribution
henryvn27
Document upstream influence, provenance, wrapping relationships, and notice obligations with explicit categories and GitHub-visible records.
SkillCFreshness: C★ 5Claude Code +9 - 0.526score
orca-auto-update
henryvn27
Model ORCA updates as a controlled channel-aware system with verification, staged rollout, compatibility checks, and rollback readiness.
SkillCFreshness: C★ 5Claude Code +9 - 0.523score
orca-context
henryvn27
Resolve ambiguous recent-work references by identifying the actual artifact, app, or screen context first.
SkillCFreshness: C★ 5Claude Code +9 - 0.523score
orca-adaptive-guidance
henryvn27
Provide lightweight, skill-sensitive coaching without slowing down normal ORCA Framework execution.
SkillCFreshness: C★ 5Claude Code +9 - 0.523score
orca-controller-mode
henryvn27
Orient a controller agent to project state, choose a path, and keep delegation and ingestion coherent across harnesses.
SkillCFreshness: C★ 5Claude Code +9 - 0.523score
orca-core
henryvn27
Apply ORCA Framework's core operating principles, artifact discipline, verification policy, and context rules.
SkillCFreshness: C★ 5Claude Code +9 - 0.522score
orca-accounting
henryvn27
Record time, retries, and optional token, cache, and cost metrics for ORCA Framework workflows without fabricating precision.
SkillCFreshness: C★ 5Claude Code +9
Signals
- Stars
- 5
- Forks
- 1
- Open issues
- 0
- Last commit (this folder)
- Jun 3, 2026 (4 months ago)
- Last release
- Jul 18, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/henryvn27/orca-framework
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from henryvn27
- 0.613score
orca-build
henryvn27
Implement planned changes with scoped edits, verification, and respect for existing project patterns.
SkillAFreshness: A★ 5Claude Code +9 - 0.609score
orca-context-brief
henryvn27
Create bounded context packets for briefed QA and layered retesting.
SkillAFreshness: A★ 5Claude Code +9 - 0.609score
orca-blind-qa
henryvn27
Run a zero-context first-look QA pass and produce a realistic first-impression report.
SkillAFreshness: A★ 5Claude Code +9 - 0.529score
chatgpt-ui-review-hvn
henryvn27
Run an iterative, evidence-based UI review loop by capturing responsive screenshots, uploading them to chatgpt.com in any available browser, applying harsh…
SkillBFreshness: BClaude Code +9 - 0.529score
senior-architect-hvn
henryvn27
Comprehensive software architecture skill for designing scalable, maintainable systems using ReactJS, NextJS, NodeJS, Express, React Native, Swift, Kotlin…
SkillBFreshness: BClaude Code +9 - 0.529score
route-models-hvn
henryvn27
Automatically select and explicitly assign the most usage-efficient OpenAI model and reasoning effort for agents, tasks, and task phases while accounting for…
SkillBFreshness: BClaude Code +9
Cite this page
orca-background-mode skill. Agent Mart. https://agentmart.dev/skills/henryvn27-orca-background-mode (last updated 2026-10-24)