Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Negotiate Run Budget

ASecurity

Use when a high-priority run reaches at least 90% of its budget and requests an extension. Never self-grants. Not for low-priority runs: let those exhaust silently.

35 stars
0 votes
0 copies
1 views
Added 9/12/2026
ai-agentsgo

Works with

cli

Security Analysis

A100/100

Scanned 9/12/2026

Install to Claude Code

$npx -y skills add OutlineDriven/odin-claude-plugin --skill negotiate-run-budget --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Negotiate Run Budget?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Negotiate Run Budget
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/outlinedriven-negotiate-run-budget-odin-claude-plugin/badge)](https://www.skillsdirectory.com/skills/outlinedriven-negotiate-run-budget-odin-claude-plugin)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: negotiate-run-budget
description: 'Use when a high-priority run reaches at least 90% of its budget and requests an extension. Never self-grants. Not for low-priority runs: let those exhaust silently.'
disable-model-invocation: true
---

# Negotiate run budget

## Contract

| Field | Bound contract |
|---|---|
| Trigger | A high-priority run reaches at least 90% of its budget and requests an extension. |
| Authority | Human-gated: asks the human budget owner for one bounded extension before continuing; every other write is reversible local, with version control as the rollback. |
| Side effect | One extension request output to the user; on decline the run enters WAITING_FOR_BUDGET report-only; never self-grants. |
| Done | Either one bounded extension is granted and recorded, or the run is report-only in WAITING_FOR_BUDGET; there is no silent continuation. |

## Inputs

Required:
- `run_id`: the identifier of the run requesting an extension.
- `budget_used_pct`: current budget consumption as a percentage (must be >= 90).
- `extension_request`: the absolute or percentage amount being requested.

Optional:
- `priority`: priority tier of the run, if present.

## Procedure

1. **Validate trigger conditions.** Confirm `budget_used_pct` >= 90. Confirm `extension_request` is a positive, bounded quantity (a percentage or absolute value, not unlimited or open-ended). If either condition is unmet, stop and issue no request. Done when: both conditions are confirmed or the run stops without issuing a request.
2. **Compose the extension request.** State `run_id`, current `budget_used_pct`, and requested `extension_request`. State the consequence: if granted, the run continues with the new budget ceiling; if declined, the run enters WAITING_FOR_BUDGET and does not continue silently. Do not add a second request, a deadline ultimatum, or an auto-escalation path. Done when: the request is composed with exactly the required fields and no escalation path.
3. **Await the human decision.** On grant: record the extension in the run state and allow continuation. On decline: record the decline and set the run to WAITING_FOR_BUDGET; do not continue the run. Done when: the decision is recorded and the run is either continuing with the new ceiling or stopped in WAITING_FOR_BUDGET.

## Failure and recovery

| Failure class | Result |
|---|---|
| Budget_used_pct < 90 | Stop. No request issued. |
| Unbounded extension_request | Stop. Do not request an unlimited extension. |
| User grants with no bounded amount recorded | Treated as decline. Run enters WAITING_FOR_BUDGET. |
| Request already issued for this run | Stop. Only one extension request per run is permitted. |

No rollback of a recorded decision. No silent continuation after decline.

## Output

On grant: run state updated with one bounded extension recorded; run continues. On decline: run marked WAITING_FOR_BUDGET in report-only mode; run does not continue. On invalid inputs or repeated request: no state change; no output beyond a brief internal stop notice.

Attribution

OutlineDrivenOutlineDriven
View sourceMore from OutlineDriven →
SSkills DirectorySkills Directory

Know which skills are safe — weekly.

Best new skills + every skill we flagged as malicious. From the team that scanned 103,619.

Join free

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Know which skills are safe — weekly.

Best new skills + every skill we flagged as malicious. From the team that scanned 103,619.

Join free

Related Skills

Caveman

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

1023331 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

651 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →