capability-permission-broker
Defines least-privilege capability brokering and approval tiers for agent and tool actions.
Skill by rupeshitpatekar
Chalk Labs
Need help deploying capability-permission-broker?
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/rupeshitpatekar-capability-permission-broker
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. 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/capability-permission-broker
What's inside
- Skill name
capability-permission-broker- Description
- Defines least-privilege capability brokering and approval tiers for agent and tool actions.
- License
- MIT
- Instructions
- 20 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Capability Permission Broker Use this skill to design permission policy between agents and tools. ## Workflow 1. Enumerate capabilities and risk tiers. 2. Map capabilities to roles and scopes. 3. Define approval mode by severity. 4. Define deny-by-default exceptions and escalation path. ## Required output format 1. **Capability Inventory** 2. **Role-to-Capability Matrix** 3. **Approval Tiers** 4. **Deny/Allow Rules** 5. **Audit Requirements**
From rupeshitpatekar/claude-code-leak-learnings, MIT.
More from rupeshitpatekar/claude-code-leak-learnings
- 0.429score
memory-conflict-compaction
rupeshitpatekar
Defines conflict resolution and compaction strategies for durable memory to preserve accuracy and relevance over long runs.
SkillDFreshness: D★ 1Claude Code +9 - 0.429score
architecture-multiagent-planner
rupeshitpatekar
Produces structured multi-agent architecture plans with explicit design decisions, constraints, and validation strategy before implementation.
SkillDFreshness: D★ 1Claude Code +9 - 0.429score
agentic-core-tool-first
rupeshitpatekar
Establishes tool-first architecture where the model plans and validates while tools perform concrete operations under strict permissions.
SkillDFreshness: D★ 1Claude Code +9 - 0.429score
agentic-core-memory-stitching
rupeshitpatekar
Guides memory stitching through indexed retrieval, evidence-gated persistence, and compaction rather than naive transcript accumulation.
SkillDFreshness: D★ 1Claude Code +9 - 0.426score
context-assembly-firewall
rupeshitpatekar
Enforces trust boundaries and sanitization in context assembly to resist prompt injection and unsafe instruction flow.
SkillDFreshness: D★ 1Claude Code +9 - 0.426score
evaluation-flywheel
rupeshitpatekar
Establishes continuous evaluation loops with scenario replay, regression suites, and risk-focused acceptance gates.
SkillDFreshness: D★ 1Claude Code +9 - 0.426score
failure-recovery-playbooks
rupeshitpatekar
Defines structured recovery playbooks for tool, orchestration, and memory failures with explicit escalation rules.
SkillDFreshness: D★ 1Claude Code +9 - 0.426score
human-escalation-gates
rupeshitpatekar
Defines human-in-the-loop approval gates for high-risk and ambiguous multi-agent decisions.
SkillDFreshness: D★ 1Claude Code +9 - 0.426score
model-routing-budget-policy
rupeshitpatekar
Defines policy-based model routing using quality, latency, and budget constraints with safe degradation paths.
SkillDFreshness: D★ 1Claude Code +9 - 0.426score
multi-tenant-data-boundaries
rupeshitpatekar
Defines tenant isolation, data minimization, and egress controls for multi-tenant agent systems.
SkillDFreshness: D★ 1Claude Code +9 - 0.426score
orchestrator-state-machine
rupeshitpatekar
Defines deterministic orchestrator state machines with explicit transitions, retries, and terminal failure handling.
SkillDFreshness: D★ 1Claude Code +9 - 0.426score
release-artifact-hardening
rupeshitpatekar
Hardens packaging and release pipelines to prevent accidental source, secret, and debug artifact exposure.
SkillDFreshness: D★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Apr 4, 2026 (7 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 rupeshitpatekar
- 0.429score
agentic-core-memory-stitching
rupeshitpatekar
Guides memory stitching through indexed retrieval, evidence-gated persistence, and compaction rather than naive transcript accumulation.
SkillDFreshness: D★ 1Claude Code +9 - 0.429score
agentic-core-tool-first
rupeshitpatekar
Establishes tool-first architecture where the model plans and validates while tools perform concrete operations under strict permissions.
SkillDFreshness: D★ 1Claude Code +9 - 0.429score
architecture-multiagent-planner
rupeshitpatekar
Produces structured multi-agent architecture plans with explicit design decisions, constraints, and validation strategy before implementation.
SkillDFreshness: D★ 1Claude Code +9 - 0.429score
memory-conflict-compaction
rupeshitpatekar
Defines conflict resolution and compaction strategies for durable memory to preserve accuracy and relevance over long runs.
SkillDFreshness: D★ 1Claude Code +9 - 0.426score
agentic-core-orchestrator
rupeshitpatekar
Defines orchestrator-first execution for multi-agent systems with explicit planning, delegation, and state transitions.
SkillDFreshness: D★ 1Claude Code +9 - 0.426score
agentic-core-stateless-llm
rupeshitpatekar
Enforces stateless-LLM architecture where execution state lives in orchestrator, tools, and managed memory layers.
SkillDFreshness: D★ 1Claude 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 capability-permission-broker 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
capability-permission-broker skill. Agent Mart. https://agentmart.dev/skills/rupeshitpatekar-capability-permission-broker (last updated 2026-10-24)