human-in-the-loop-promotion
Review policy, evaluations, cost, lineage, and risk before recording a human promotion or rejection decision.
Skill by closedloop-technologies
Chalk Labs
Need help deploying human-in-the-loop-promotion?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
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/closedloop-technologies-human-in-the-loop-promotion
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
Use cases: agent-building
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 and code. Checked Nov 1, 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/human-in-the-loop-promotion
What's inside
- Skill name
human-in-the-loop-promotion- Description
- Review policy, evaluations, cost, lineage, and risk before recording a human promotion or rejection decision. Use when a ClimbHill attempt is ready for disposition.
- License
- MIT
- Instructions
- 39 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Human-in-the-Loop Promotion Use this skill when an Attempt is ready to be promoted, rejected, or converted into follow-up work. ## Inputs - Attempt comparison - Attempt patch or branch - Evaluation records - Policy check results - Cost summary - Risk notes - Maintainer decision ## Procedure 1. Confirm the attempt has no denied path edits. 2. Confirm approval-required paths have explicit approval. 3. Confirm required tests and evaluations passed. 4. Review cost, risk, lineage, and report summary. 5. Ask the human to choose promote, reject, request changes, or run another loop. 6. Record the decision with actor, rationale, run ID, and attempt ID. 7. If promoted, prepare a PR-ready branch and checklist. 8. If rejected, record reasons so future attempts can avoid repeating the failure. ## Outputs - Human decision record - Promotion or rejection rationale - PR-ready checklist or follow-up issue proposal ## Rubric - No promotion happens without a recorded human decision. - Policy and evaluation status are visible. - Rejections preserve useful lessons. - The next step is clear and reversible.
From closedloop-technologies/climbhill, MIT.
More from closedloop-technologies/climbhill
- 0.584score
historical-attempt-sampling
closedloop-technologies
Sample successful and failed historical ClimbHill Attempts before planning new work.
SkillBFreshness: B★ 7Claude Code +9 - 0.584score
prepare-codebase-for-recursive-optimization
closedloop-technologies
Prepare a repository for controlled recursive optimization by defining goals, agent guidance, policies, evaluations, resources, and protected surfaces.
SkillBFreshness: B★ 7Claude Code +9 - 0.584score
protected-surface-editing
closedloop-technologies
Classify proposed paths and enforce denied or approval-required ClimbHill policy before edits.
SkillBFreshness: B★ 7Claude Code +9 - 0.584score
resource-grounded-coding
closedloop-technologies
Ground implementation decisions in recorded ClimbHill resources and cited prior work.
SkillBFreshness: B★ 7Claude Code +9 - 0.584score
evolving-solutions
closedloop-technologies
Recombine proven parts of multiple evaluated attempts into a coherent follow-up attempt.
SkillBFreshness: B★ 7Claude Code +9 - 0.581score
meta-analysis-to-github-issues
closedloop-technologies
Convert evidence-backed ClimbHill experiment findings into reviewable GitHub issue proposals.
SkillBFreshness: B★ 7Claude Code +9 - 0.581score
experiment-tree-analysis
closedloop-technologies
Analyze ClimbHill runs, attempts, lineage, costs, failures, and recurring constraints.
SkillBFreshness: B★ 7Claude Code +9 - 0.581score
cost-and-progress-reporting
closedloop-technologies
Produce evidence-backed run and attempt reports with evaluations, costs, risks, and promotion recommendations.
SkillBFreshness: B★ 7Claude Code +9 - 0.581score
parallel-coding-agent-loop
closedloop-technologies
Coordinate independent, bounded coding attempts against one goal and compare them after policy and evaluation checks.
SkillBFreshness: B★ 7Claude Code +9 - 0.539score
deep-research-perplexity
closedloop-technologies
Run Perplexity Sonar and Sonar Deep Research for web-grounded answers with citations and bounded search context.
SkillCFreshness: C★ 7Claude Code +9 - 0.539score
deep-research-custom-data
closedloop-technologies
Build and validate custom-corpus inputs for deep research from Google Docs, YouTube, local files, S3, arXiv, databases, Slack, email, tickets, and similar…
SkillCFreshness: C★ 7Claude Code +9 - 0.539score
deep-research-exa
closedloop-technologies
Run Exa-backed deep research, neural search, and page-content retrieval with bounded result and page-read counts.
SkillCFreshness: C★ 7Claude Code +9
Signals
- Stars
- 7
- Forks
- 0
- Open issues
- 1
- Last commit (this folder)
- Aug 20, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from closedloop-technologies
- 0.584score
prepare-codebase-for-recursive-optimization
closedloop-technologies
Prepare a repository for controlled recursive optimization by defining goals, agent guidance, policies, evaluations, resources, and protected surfaces.
SkillBFreshness: B★ 7Claude Code +9 - 0.584score
resource-grounded-coding
closedloop-technologies
Ground implementation decisions in recorded ClimbHill resources and cited prior work.
SkillBFreshness: B★ 7Claude Code +9 - 0.584score
historical-attempt-sampling
closedloop-technologies
Sample successful and failed historical ClimbHill Attempts before planning new work.
SkillBFreshness: B★ 7Claude Code +9 - 0.584score
evolving-solutions
closedloop-technologies
Recombine proven parts of multiple evaluated attempts into a coherent follow-up attempt.
SkillBFreshness: B★ 7Claude Code +9 - 0.584score
protected-surface-editing
closedloop-technologies
Classify proposed paths and enforce denied or approval-required ClimbHill policy before edits.
SkillBFreshness: B★ 7Claude Code +9 - 0.581score
parallel-coding-agent-loop
closedloop-technologies
Coordinate independent, bounded coding attempts against one goal and compare them after policy and evaluation checks.
SkillBFreshness: B★ 7Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get human-in-the-loop-promotion running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
human-in-the-loop-promotion skill. Agent Mart. https://agentmart.dev/skills/closedloop-technologies-human-in-the-loop-promotion (last updated 2026-10-24)